Jacques, Waly- I have no problem with the diagnosis - it has been made before. Among others, Angus Deaton made it 10 years ago, but, it is great for you to raise it again. We need to keep saying this. It is a tough thing to say, but we have to say it. our excellent research department has developed the tools, but they are not being applied. So we have confusion. Several points. (1) There are several texts and books which are considered the "bibles" on the subject.. However, what is considered good practice in middle and low income countries is often not applied in SSA because we don't have the data, or the quality is poor. For example, the rental value of owner occupied housing. We probably need more survey research to figure out exactly how to apply the good practice techniques in poorly monetized economies. (2) I also think you overestimate the degree of harmonization and quality of the national accounts. As with poverty, few low income countries in SSA use the agreed methods - most are stuck in SNA of the 80s or worse. And have you seen the agricultural statistics in Africa? These are often produced with very primitive methods, despite the fact that it is actually quite difficult to estimate production when you have so much intercropping, fragmented land plots, and home consumption or barter. (3) Finally, one of the main differences between the $1.25/day and the national poverty lines is the poor quality of countries' CPI. Many of these have weights that are 20 years old. So we need to work on that as well. In other words, it is all related. it is not just poverty numbers. So I would say - overall, the quality of data produced by most low income country AFR stats office is at best iffy. We need to start being a lot more cautious about the "trends" we find. And then we come back to the basic problems of leadership and financing. What is going to be done and who is going to do it? Who will the lead this effort, and who will pay for it? In Latin America, they had a 10 year effort involving The World Bank, The IDB, and the UNECLA, called MECOVI. In those days, there was more money for this type of effort. We should have done the same thing for SSA at that time. We didn't, and we lost time. Who would pay for such an effort now? Do we have the capacity to lead such a multi-year capacity building effort any more at the World Bank? Does management have the desire, and the vision? or are we just happy when the numbers go up, and then when they go done we complain about quality of data?