There's a very good The New Yorker article www.newyorker.com/fact/cont...30fa_fact1 on micro-lending.

I first saw it linked to by Libertarian economist Tyler Cowen www.marginalrevolution.com/marg...ments who took this snippet out of the article:

"Omidyar and his colleagues say that the biggest obstacle to commercialization of the sector is philanthropic capital. They say that it distorts the market—not only by filling channels that might otherwise draw commercial investors but also by keeping unsustainable programs alive."

Pierre Omidyar has the eBay fortune, and is interested in social entrepreneurship. His point is not to be taken lightly. But business in the developed world rests on institutions that often don't exist or are poorly developed in the developing world, so it matters very much what the programs and associations which enable micro-lending operate. Bill Gates, from the sounds of the article has concluded that there are many structures that can support micro-credit and micro-lending and is not as worried as Omidyar about philanthropy spoiling a good thing.

www.trickleup.org/ Trickle Up supports small grants for the very poor and is just one example of enterprises as an approach for poverty reduction.

Banking services for poor people involve more than access to credit. One of the issues I've been looking for is ways that people in the developing world can get paid for services. For example people in Uganda could do data entry for clients in the USA, but whereas people in the USA can do that and get paid via paypal no such services exist in Uganda. Here's an interesting paper on mBanking www.infodev.org/en/Publication.169.html

Jan Chipchase travels the world for Nokia to discover uses for their mobile phones they haven't thought of yet, his blog is quite cool www.janchipchase.com/blog/ar...hannels/ He cam across a system for sending money called Sente. Mobile phones are charged, but the value is turned over to a middle man who transfers the money in a network for a commission. The middle man then sells the phone time per minute to recover the money--explained much better here networks.silicon.com/mobile/...3,00.htm

You raise a very good point, Lazarus about the inefficiency of bureaucracies, but businesses for the poor need more structures than simply access to credit in my opinion. Now the innovation needed is to put into play some of the innovations that organizations like Kiva have made to include peer to peer production into play to build the organizations which can help poor people build businesses.