Thanks for the comment. I think you are absolutely right: Information in and of itself has a direct impact only when people either have options or can act on the basis of that information in other ways. The emerging research on this is converging quite rapidly to that idea. For instance, our experiment of providing report cards on schools and children in Pakistan reduced prices in private schools and increased test-scores in both public and private schools. At the same time, it's worth emphasizing that the effect of providing new information depends on the extent of information that was already available in the system, and there are few systems where the baseline was such that no information at all was available. Again, in the Pakistan report cards experiment, we are able to show that the new information increased the precision with which people were able to assess schools--but even in the baseline, their perceptions were quite accurate.
I am sure Shanta will have more to say about this, but its worth emphasizing that information also plays the key role of bringing something on to the legislative agenda. I don't think that everything that is backed up with hard information gains legislative attention, but hopefully, the likelihood that it does increases. I think absenteeism and the lack of incentives among public sector workers gained increasing acceptance as a problem after the work on the WDR, and we have seen some really innovative programs come up to deal with these.
So, my take is
1. Exactly as you say, more information leads to better outcomes where people can take actions on the basis of the information
2. New information also provides a means to move away from discussion based on experience towards an agreed upon set of facts. This in turn could--and has--created a platform for policy change in some cases.