This is a great idea. Giving directly to the poor will hopefully bypass corrupt entities, and essentially injects cash directly into the economy you're hoping to assist. In the case of a farmer or someone looking to start a business, even a small cash injection can make a big difference, and will immediately produce an impact.
The main risks of course to this plan is that the money could get spent on imported goods, which would transfer the cash back out of the local economy, or that it would get hoarded (or stolen). But cash injections are still the most effective means to stimulate an economy, especially small, local economies.
"cash injections are still the most effective means to stimulate an economy, especially small, local economies."
The economist in me thinks the effectiveness of cash stimulus is still mostly inflated away if the economic problem is structural, especially when injections are unrelated to any productive improvements in small, local economies. To the extent this scheme works because the cash injections are occasional and unpredictable and landlords, local Big Men and monopoly local goods suppliers haven't yet found a way of creaming off most of the cash injection. I wish them all the best and hope the positive results their followup studies have shown reflect something other than weaknesses in study methodology, but unfortunately I think this is patching the symptoms and can't scale without adding in some more precisely-targeted larger-scale aid.
The main risks of course to this plan is that the money could get spent on imported goods, which would transfer the cash back out of the local economy, or that it would get hoarded (or stolen). But cash injections are still the most effective means to stimulate an economy, especially small, local economies.