One idea that came out of the last conference was that it made sense for us to develop a common project for various groups in attendance, and that for this project it made sense to identify a key issue that we all address and that we might want to measure our different governments against some type of common standard. The belief was that a joint project was not only a good way to share knowledge if we have the right theme, but also that the work in the different countries might reinforce each other.

This led to the idea of the transparency scorecard. Those of you on the list serve may remember the exchange that we had about it a few months after the conference. Let me begin with the reasons why we thought a transparency scorecard might be a good joint effort. The first is indeed it is a common challenge that we all face. Second, we thought that it was a very important starting point for all our work. As we all know, the lack of transparency is a necessary obstacle to overcome before we can do our work. We can't do analysis without information and we can't put that analysis into the public debate unless the debate is constructed in a way that allows for participation.

So the initial idea presented to the List serve was to develop something I had labeled a Transparency and Democracy Scorecard, to rate a number of countries against it, and then to do some type of simultaneous release around the world that would include individual reports for a number of countries as well as a general report synthesising the results.

This idea subsequently has been modified for two main reasons. The first reason had to do with the List serve discussion itself, which raised several important points that we had to take into consideration. One point was that an emphasis on democracy wasn't one that everyone felt comfortable with and that the concept of participation was closer to what people thought should be incorporated into the report. Democracy and participation, of course, are closely linked.

The second point raised by the list serve discussion was that it was not clear what the right scoring system would be. Should we give countries a grade or would that be too simplified and not allow for the nuances and subtle differences between the countries? There was also a concern that it didn't make sense to compare countries against each other because that doesn't allow for an analysis of the context that the different countries face. For instance, should the same set of transparency standards apply to the U.S. as apply to South Africa when the U.S. has had more than 200 years to develop an open budget process, and South Africa has just had the experience of 4½ years to develop such a process?

The second major reason that we decided to modify the initial idea is that we came across the IMF standards after we had discussed this on the List serve, and the standards do provide an excellent base from which we can depart and from which we can work from. We realised that we should and had to take the IMF standards into account. So the idea we came up with was that instead of launching the project with the simultaneous release approach, when there were all these issues to sort out about what is the best scorecard and how it can be applied, we decided to start with a pilot effort.

This is the effort that we have been engaged in with Idasa. A pilot effort allows us to develop a methodology. It allows us to assess all the questions that were raised in the List serve discussion and it allows us to figure out what our approach should be relative to the IMF model. Based on this report and experience I think we can really figure out how to proceed and that hopefully there will be a lot of discussion on what to do — so not only this session but this afternoon is specifically designed to discuss those issues.

One of the ways we can proceed, of course, is to say, well the model we developed is pretty good. Some of you might want to take the exact model and run forward with it. My guess though is that we will have a few modifications at least. We may ultimately decide to do some type of cross-country synthesis. A cross-country synthesis will be a lot easier if we are working from essentially the same model and the advantage of the cross-country synthesis is that it makes it easier for the work to reinforce each other.

We might also decide to develop specialised reports. One such request was raised by Richard Humphreys and others in the question period of the last session, and that is we can develop a scorecard that is particularly applicable and appropriate to the sub-national level. The scorecard that we used did shed light on issues at the sub-national level in South Africa, but there are some different questions that we would ask if we began at the sub-national level and decided to develop a scorecard for those issues.

We might also develop a military transparency scorecard. When it comes to transparency it is hard enough to achieve for domestic issues, but of course it is much more complicated and tougher to accomplish in the area of military expenditures. So there might be some basic defense budget questions that could develop into a separate scorecard or perhaps that could be incorporated into our survey, which we didn't do at this point.