The US government learned many hard lessons from its military occupation of Iraq. With respect to corruption in security and reconstruction projects, one of the clearest lessons—emphasized by the 2013 final report from the Special Inspector General for Iraq Reconstruction (SIGIR), among others—was that smaller, short-term projects were more effective, and less susceptible to massive and debilitating corruption, than big, long-term projects. Indeed, a month after publication of the SIGIR report, Paul Cooksey, the Deputy Special Inspector General for Iraq Reconstruction, testified to Congress that large amounts of money should not be injected into an unstable region without enough well-trained, experienced personnel to oversee it. The better strategy, he argued, was to use small projects that could be more tightly managed. For example, one battalion commander in Iraq mentioned that greenhouse and drip irrigation projects—which allowed farmers to use water more efficiently and grow vegetables year-round—were small enough to be easily monitored to completion. This may not be as grandiose as building massive infrastructure, but it can still have a meaningful impact on people’s lives.

Yet despite the clarity and consistency of this message, it has not been heeded in Afghanistan. When the U.S. invaded, there was an accompanying flood of aid. General Stanley McChrystal championed pouring money in, and General David Petraeus saw the ability to dole out cash as an important political weapon to build security on the ground. The result is that U.S. foreign aid averaged $15 billion a year, making Afghanistan the highest recipient of official development assistance in the world for several years. The Afghan government depends on aid for about 70-90% of its budget. Much of this money went to large-scale projects. And, much as one might have predicted if one paid attention to the retrospective evaluations of the Iraqi reconstruction, these large projects fed a cycle of corruption: the US (or the Afghan government, with US money) would give substantial amounts of money to a large contractor, which would in turn subcontract to a smaller company, which would subcontract to an Afghan NGO, which would then hire a local contractor to do the work. Money leaks out at every stage, a corrupt actor usually ends up with the contract, and bribes are paid along the way to local officials, police, and even the Taliban. All of this can increase the cost of projects by a factor of ten, and aggregated across the economy, this massive spending can lead to distortionary effects.

An example this dynamic at play comes from the Overseas Private Investment Corporation (OPIC), an independent U.S. agency responsible for development financing. OPIC loaned out tens of millions to fund hotel and apartment projects in Afghanistan, with the vision of a five-star Marriott Kabul Hotel. Money was given to the developer, but OPIC failed to ever visit the site or verify claims that the developer made. OPIC did hire an independent contractor to oversee the project, but the contractor also failed to visit the site or verify claims. Nearly a decade after the project was first conceived, the Special Inspector General for Afghanistan Reconstruction concluded that the money had disappeared, the buildings were uninhabitable, and the developers appeared corrupt. This is not an isolated example: of the many large infrastructure projects supported with aid money in Afghanistan, many have since deteriorated due to poor workmanship—to which corruption is undoubtedly a contributing factor.

Smaller projects, by contrast, are less susceptible to corruption. When projects are smaller, there is less need to retain a wave of subcontractors, and it is thus easier to ensure that there is adequate oversight and management. This is a critical point, as lack of oversight has been identified as a key driver of corruption in Afghanistan. Nearly a dozen different agencies have criticized the poor oversight of contractors, which opens the door to corruption.

In order to effectively oversee a disbursement of money, there must be a well-trained team of staff with experience handling these sorts of projects, known as contracting officer’s representatives (CORs). The Government Accountability Office has noted that many CORs receive inadequate training and are understaffed. This means they are incapable of scrutinizing complex technical aspects of projects, and may be given far more work than they can reasonably handle. Focusing on small projects could help alleviate the workload, as small projects are less complex. The number of projects should also be reduced to a level the military can adequately staff, even if that means moderating the level of aid that enters Afghanistan.

To be clear: I am not calling for an end to all foreign aid. It has a valuable role to play in humanitarian efforts and in building security, and there have admittedly been some positive impacts from large-scale aid spending in Afghanistan. But on the whole, if a government does not have the capacity to process and distribute foreign aid, throwing more money at the problem without a coherent strategy will not do the people any good. Smaller projects would be more effective in the short-term and if done right, could help make it easier for the U.S. to smoothly exit the country in the long-term.