What is missing in all the postings is an understanding of the politics of public administration in African countries. This is essential if we are to determine what drives the public adminstrator's performance, since this is what we want to influence in order to improve the overall performance of the public sector. Examples in the previous blogs include the concept of "nascent" governments and "established" governments. Even post-conflict countries such as Sierra Leone and Liberia have well established civil services with traditions and survival systems that are well honed, albeit in corrupt practices. Unless we understand underlying forces determining behaviour in these administrations we are bound to continue to think in terms of capacity development in the form of training, computers and cars. The solutions proferred by donors are even worse. As stated by Nils Boesen, donors are excurciatingly slow at learning from earlier errors. Again understanding the politics of donor behaviour will help us in understanding why. Here we all know that the average donor's reaction to cover up and justify becomes a natural reflex after only two to three years on the job. So they keep repeating the same errors and explaining it away by pointing at the incompetent and corrupt beneficiaries. At least Sam Gardner seems to be on the right path by pointing out that the true explanation is not the lack of degrees, and I would add in most cases, but poor priorities. Graham Teskey scratches the surface but does not go deep enough. I will offer one hint; what keeps a public official in his post is less performance than connections. Donors cannot eliminate the latter and replace it by the former. Only local efforts and time can achieve this. In the meantime donors must understand this and think outside the box to link the two. The lower the country is in the development ladder the more difficult this task.