A bit more than 300$ went into Kiva donations (you can choose for every loan, to cover administrative costs). 20$ went to currency losses. 80$ went to default losses.
The default losses are 10% of what I put in, or 1.6% of the total amount lent. And I explicitly chose a few loans that were explicitly marked as high-default-risk, because of the social situation of the recipients or due to the political/economic situation in the region.
I cannot confirm that my money dried up in default losses. I still think it's quite a good model (but I haven't followed economic research about microloans in the last few years).
Edit: I grew up in the amazon region, parents worked in an NGO. Giving people gifts/donations is worthless to most, since something gifted feels like it has no value. If you build a water well for free, it will be broken in a few months to years. If you make the village pay a part of the costs on the other hand, and if they must help building it, they'll value it more and keep care of it. It's a cultural thing. I could imagine that loans work the same way: You don't get a donation, you'll get a loan that you have to work for. In that culture, it will be valued more. (Of course, that applies to the region I grew up in, I don't want to over-generalize.)
100 dollars have been used for 500+ loans, the last 8 years or so. While you can pick the recipients yourselves, I've opted for automatic re-lending on the platform.
That really depends on how much they want the security. People defend territory all the time. It's very, very expensive to do that -- have a look at some city walls -- but it's cost-effective because they are using the territory.
But no, no matter how large the group, they cannot buy cost-effective security for their giant stretch of wilderness if their plan for it is "don't use it" and there are other people who do have a use for it. Compare American government efforts to stop people from using land to grow marijuana. Is the problem that the American government just isn't big enough?
People are generally doing that in areas where people are allowed. Also the government has a stupendously large amount of land that it didn't pay for and a low budget for keeping anyone out.
It's different if this is land you're purchasing in the modern age. If you're buying a million acres for over a hundred million dollars, then that's about 150 miles of border to patrol. If you hire 150 people for that, at above-median wages, that's 3 million a year. Add some overhead, some expense for equipment and buildings, it's affordable enough compared to the purchase price. If that's too expensive then go up to 10 million acres, which triples the land area per mile of border. Or 100 million acres.
> If that's too expensive then go up to 10 million acres, which triples the land area per mile of border. Or 100 million acres.
I don't really follow. The use case is that you want to prevent small farmers from encroaching on the jungle. This will only ever happen at the border between where the small farmers end and the jungle begins. If you buy land behind the border, that's nothing but a waste of your money (until the border is so thoroughly dejungled that your old hinterlands is the new border).
In turn, this means that all of the land you're interested in is border. You can't rely on the fact that a circle has a high ratio of area to perimeter to claim cost-efficiency, because you're structurally constrained to owning maximally cost-inefficient strip shapes.
But assuming your scenario is the right one:
Buying only the border land is a strict improvement in efficiency over my plan. Instead of paying X million for land and Y million for security to protect a huge swath of rainforest, you only have to pay X/50 for land and Y million for security. Who cares about what percent goes to what, you're saving tons of rainforest for even less money.
In the extreme you might realize you don't even need to buy any land, you're just paying for security to protect the rainforest. Instead of paying 100 billion to own and never visit, you pay 1 billion a year to protect land you don't own. Which... sounds fine to me!
You probably need a watch tower about every 1~2 km, or even less if you'd use a strip of minefield to protect against hordes.
There's a similar problem with your idea. You could buy a stretch of land in Brazil and bask in the glow of taking it off the market. But the problem you perceive with overuse of the Brazilian rainforest wouldn't be solved; that land was mostly not on the market anyway. How would you stop farmers from encroaching on your land? A "No Trespassing" sign won't do the job.
Buying land to prevent development will work better in the US, mostly because there's no one in the US who lives off the land. You can't stop small farmers from moving on to your unsupervised land. But you can easily stop giant logging or mining companies from doing it.