GAB is delighted to welcome back Mark Pyman, Senior Fellow at the London Institute for Statecraft, who also served as Commissioner of the Afghanistan Joint Independent Anti-Corruption Monitoring and Evaluation Committee until November 27, 2017.

When it comes to fighting corruption and promoting accountable government, donors provide funds, expertise, and support, often over many years. They face many difficult challenges, and we all sympathize with the hard issues they have to contend with. Yet at the same time we have to forthrightly acknowledge that, for all their good intentions, when it comes to corruption, international donors easily become part of the problem. Donors, researchers, politicians and grantees have all been too silent on this.

Let me illustrate this with problems at one large, well-intentioned donor program in Afghanistan, the Comprehensive Agriculture and Rural Development Facility (CARD-F) Program. This Program, funded by the UK’s Department for International Development (DFID) and Denmark’s aid agency DANIDA to the tune of $120 million over two phases, was established to increase rural employment, incomes, and business opportunities through the design and implementation of projects, such as infrastructure work (such as building irrigation canals), provision of grants to producers and processors, establishment of greenhouses and poultry farms, and training for farmers.

Between March and October 2017, the Afghanistan Independent Anti-Corruption Monitoring and Evaluation Committee (MEC) made an inquiry into corruption concerns at CARD-F, based on allegations from five whistleblowers. MEC is the premier anti-corruption entity in Afghanistan, set up by Presidential decree in 2010, led by a Committee of six (three eminent Afghans and three international experts), and with an Afghan Secretariat of some 25 professional staff. It is funded by international donors. MEC found plenty of malpractice, including nepotism and cronyism in the Management Unit; multiple irregularities in the awarding of grants and procurement contracts; poor monitoring provided by expensive UK companies (that, to be blunt, were not doing their job); and international (UK) contractors with a built-in incentive to use up more of the available budget for their own “technical assistance.” MEC found that only 33% of CARD-F funds in the first phase reached the intended end users, instead of the planned 60% (the other 40% planned to going on technical assistance and administration; eventually 67%). Moreover, not one of the five separate whistleblowers whose concerns were passed to MEC felt protected enough to complain through the CARD-F program, nor through DFID or DANIDA. At least two of these whistleblowers were fired, and others felt they had to leave.

At the same time the donors vigorously opposed MEC’s plan to do the inquiry, suggesting that MEC surely had other more important priority topics to examine, and that MEC shouldn’t be concerned because the donors had already done an audit (which was not shared with MEC) in response to a previous whistleblower. Not-so-subtle pressure was applied: MEC’s own core funding, which comes partly from DFID and DANIDA, would need to be “reviewed” if MEC persisted. Ultimately, MEC had to request the President of Afghanistan to intercede, before DFID Afghanistan offered its support to MEC’s inquiry.

Any organization doing or sponsoring work in a tough environment like Afghanistan can expect to have corruption issues. But trying to hide the problem, and then to bully it away? As an anticorruption professional who has seen DFID do good work elsewhere in the world, and indeed in Afghanistan, I was really shaken. Less naïve than me, the Afghans are well aware that such internationally sanctioned malpractice is taking place, and they too see this as evidence of dishonesty and hypocrisy.

The huge disconnect between donors’ generally good intentions on the one hand and the, frankly, perverse bureaucratic politics that drives donor agencies is a known problem. Most donors know what is going on in their programs, but feel driven to cover themselves with expensive and often ineffective technical veils – fiduciary risk assessments; supply chain mapping, due diligence, layers of oversight – to protect themselves from charges of lax supervision.

An honest conversation about this is surely overdue. Here are ideas on four of the key topics to start the discussion:

1. Becoming captured by the cause of doing good and over-identifying with the mission. It is almost ubiquitous to find sector experts worrying that discussing corruption might alienate the “client” nation, and/or might lose them financial support in their home Congress. This is pernicious, but the system incentivises this behaviour. Donor funding is highly compartmentalized, with funds for sector support, such as health, education, or infrastructure, and usually a separate funding arrangement for “governance” support, which includes anticorruption. Domain experts, e.g. in agriculture or education, know that there will not be support for anticorruption work within the sector funds, so deliberately do not pursue it further, even when they wish to. USAID, for example, reviewed 330 worldwide projects of value $6.7 billion during 2007-2013. Of the anticorruption components of the projects, 74% was for general “governance” support; only 1.7% was specific to health, for example. In these types of sector work, such donor corruption concern that exists is not focused on strengthening the system, but narrowly focused on protecting their own aid money.This can be at least partly addressed by changing the donor funding structures, so that sector-specific anticorruption projects can also be programmed within the sector funds.
2. “Monitoring” does not work well and different local solutions need to be fostered. There is a real, substantive need for better monitoring mechanisms in difficult countries, and donors need to re-visit both the options and the risk/corruption trade-offs. On the one hand, most of what CARD-F does could have been done by normal government programs, which would have been a lot more sustainable but also, in principle, riskier. Here DFID got the worst of both worlds – neither better controls nor more sustainability; as did the country, with only 33% of the funds ending up in the hands of the poor. On the other hand, there is scope for other local solutions, such as setting up NGOs with a specific local monitoring brief. Look, for example, at how SIGAR extended its monitoring in-country by using the NGO Integrity Watch Afghanistan as a local monitoring agent.
3. International monitoring is half-hearted, and is too much under the control of the local country director. The program monitoring that is done for development agencies, usually on a contracted-out basis or by auditors, is not independent but is under the control of the Agency’s Country Director. This gives too much power to the Country Director. Contractors are very alert to unspoken messages about how not to embarrass head office by finding corruption problems. In Afghanistan, for example, several of the international donor audit teams have never found any cases of corruption in their programs, which rather defies belief. Similarly, a 2012 official analysis of DFID projects in Afghanistan comments that “organisations in the delivery chain have a vested interest in reporting success…most organisations we interviewed commented that frauds and corruption were either non-existent or trivial within their programmes.” Whilst it makes sense that the main monitoring contract should be established and run locally, the donor head offices must have a robust capability to monitor the monitors. Such a lack of accountability is immediately recognised locally. It allows ‘over-protective’ donor staff to cross the line to really unacceptable behavior, such as not taking whistleblowers seriously and allowing them to be fired without their complaints being followed up; and it means that national staff are only too aware there will never be a genuine complaints system with certain Country Directors in place. This problem exists both for assistance from nations and multilateral assistance, such as from the World Bank.
4. Technical assistance is often of unprofessional quality, but can rarely be challenged. If there was proper discussion, donors would likely admit that a significant proportion of the technical assistance is of very little help and should be being weeded out. Shouldn’t there be tougher performance requirements on technical assistance? Sanctions for contract non-performance in off-budget programs?

I welcome your feedback: between us we can surely initiate a new conversation on this.