Perspectives from Implementing Countries on Technical Support Issues
Two months ago, Aidspan's Executive Director visited four countries in Asia and Africa to discuss technical support issues with people who are involved in or supporting Global Fund grant implementation. People were asked to speak candidly and privately, so that their views could be used for a document containing various anonymous "perspectives from implementing countries" that would be provided to participants at last month's Round Table meeting on technical support (see previous article).
This article contains a summary of the resulting document, which is made up of different (and sometimes conflicting) points made by various individuals in the course of the above discussions.
- The term "technical support" (TS) is used at different times to cover different activities. These include: providing policy or technical advice in a difficult or new area; providing management advice or support; providing coordination advice or support; and providing capacity building, which in turn is used to mean training new staff, and/or providing additional training to existing staff.
- Effective progress with GF grants (and with other such initiatives) requires all four legs of the following table to be strong in-country: Leg 1: Technical/management skills and capacity. Leg 2: Effective coordination among the many players. Leg 3: Political leadership/support. Leg 4: Absence of endemic corruption. If any one of those four legs is weak, the table runs the risk of collapsing. TS can sometimes address weaknesses in Leg 1, and can occasionally address weaknesses in Leg 2, but it cannot deal with weaknesses in Leg 3 or 4.
- You can't use TS to solve an unsolvable problem. For instance, when there is a government Principal Recipient (PR), and substantial portions of the grant are supposed to be passed by the PR to Sub-Recipients (SRs) outside government, yet the government is not firmly convinced of the need for SRs outside government, there will inevitably be major problems or delays or blockages, and no amount of TS will resolve the issue.
- For each GF grant, there can be a need for TS in five areas: Area 1: Develop the proposal. Area 2: Go from the approved proposal to an implementable workplan. Area 3: Get the CCM / PR / SR players to agree on how to work together. Area 4: Deal with purely technical/programmatic issues like how to scale up testing, how to do procurement, etc. Area 5: Meet the GF's accountability requirements (re M&E etc.). These areas require very different types of TS. Governments have been perfectly willing to seek TS in Area 1. But once the grant has been approved, governments have often felt that the money is "theirs", and have been reluctant to seek TS in the other four areas, and have been unwilling to take actions which could lead to a reduction in their control over the money.
- The proposal submitted to the GF often doesn't have an accurate self-assessment of TS needs. By the time the need has become clear, and the money found, and the consultant found, it's often too late.
- In many countries, GF grants would work much more effectively if much of the PR's work was outsourced to a Project Management Unit (PMU), probably in the private sector, that was given a Service Level Agreement (SLA) which specified deliverables and targets - e.g. that disbursement requests must be submitted to the GF, and approved, by a specified date. The PMU would be allowed, when it met its service level targets, to charge a management fee that was a small percentage of the grant. In this context, the PMU would be much more likely to request TS when it was needed, because the PMU would see that receiving appropriate TS could enhance its chance of earning its full management fee. Sophisticated corporations have no qualms asking for help from consultants. Why can't PRs be the same?
- All too often, TS providers provide the TS that they can provide, rather than the TS that is actually needed and/or wanted. And almost everyone ends up happy: The TS provider earns money; the funder of the TS looks good; people can assure the GF that there is forward movement; and anyone who dislikes the recommendations can ignore them. Of course, if this means that the recommendations are ignored by the CCM or PRs or SRs or whoever needs to take action, what you have is lots of appearance of movement but no real movement. Which means that the one party that does NOT end up happy is the people who need to receive services as a result of the GF grant.
- The main problems with TS have been: Too much ad hoc consulting; too much supplier-induced consulting; international consultants who come in with wrong credentials and/or inadequate briefing; recipients not knowing what the consultant is doing or why the consultant is there; international consultants working with the wrong kinds of people in-country; and international consultants who lack skills in communicating with their local counterparts.
- Recipients of TS have a responsibility to lead the consultant and to provide feedback when the consultant's work is not good enough. But this rarely happens. Indeed, fifty percent of the reports of international consultants aren't even read - let alone acted on - by some recipients, which makes those reports a complete waste.
- Part of the new mantra is that TS should be "demand-led" rather than "supply-led". The reason it shouldn't be supply-led is that the "supply" party doesn't fully understand, and certainly doesn't control, the situation at the "demand" end. But the problem with demand-led TS is that the "you don't know what it is that you don't know" - meaning, the party at the demand end doesn't necessarily know what are the skills that he/she lacks and that could be provided by the party at the supply end.
- There is a need for "proactive coordination". That is, some party (usually, an international person based in-country) should: (a) Get domestic parties to articulate what the demand for TS actually consists of. (b) Ensure that TS is provided in a way that meets that demand. (c) Persuade domestic parties to implement what is recommended by the TS provider. The momentum will dissipate if there isn't someone playing that role.
- When a bilateral or multilateral donor gives a grant to a country, and then that donor also provides or finances TS (whether or not the TS is related to the grant), the recipient is very hesitant to reject or criticize the TS, for fear that this will also threaten future grants from that donor. And that creates a mentality of mute acceptance regarding donor-provided/funded TS.
- When the grant implementer is weak and/or unclear, it's very hard to know how to provide appropriate TS. The grant implementer won't ask for it, and the TS provider/funder can't really know what is needed; and even if they think they do, the chances of it being accepted are not great.
- Certain UN officials in some countries spend up to half their time dealing with GF-related issues. They serve as the "oil in the machine" or as "proactive coordinators". This is a need - and a cost - that needs to be recognized. Without these officials playing that role, the machine (i.e. the running of the CCM and the grants) runs the risk of slowing down and becoming much less effective.
- There is a need, within each country and regarding each disease, for TS to be coordinated by one agency (the "lead agency"). This doesn't mean that the lead agency has to provide all the TS, or to control who provides it. But the lead agency should look for and seek to avoid gaps and overlaps in TS provision. And it should be accountable, so if TS is not provided or is ineffective, people know who to point the finger at. However, the choice of who is made lead agency should be repeated each year, based on performance, and should be based as much on personal strengths as institutional ones.
- There are huge problems with loss of institutional memory. Just as someone (civil servant, UN official, etc.) gets to the point of understanding what's really going on, they get reassigned, and the memory is easily lost.
- Someone should develop a menu of types of TS that may be needed in different circumstances, and for each type, should develop, and make widely available, a template of the TOR document that others can then draw upon.