I think that's a really interesting middle ground. Do you think it can fall into the same trap as mentioned in the article, where it can become "sub-optimal" due to not being as expert as the non-profit?
It's exactly the problem in the article, taken to a ludicrous extreme. If you want to be able to decide how a nonprofit should spend their money, go take a seat on the board (or staff).
I think it’s less prone to that problem if you take a holistic approach when evaluating whether a loan should be repaid, rather than applying very narrow criteria such as “must be spent on a new building”.