It does require more data and analysis, but is entirely possible. I'm not taking a stance on this (in reference to your 'dubious' comment) but I'm trying to ela

It does require more data and analysis, but is entirely possible. I'm not taking a stance on this (in reference to your 'dubious' comment) but I'm trying to elaborate on how this would be implemented. Consider a database where each account has the owners name, local address, other demographic data in addition to their account balance. The database also stores their transactions. Now allow businesses to have their own central bank accounts as well. The business has its location and type of business listed as well. Assuming all this data is entered correctly and up to date the implementation of "$100 to be spent in 30 days on local restaurants" is as follows. Assume local is, within 10 miles of the account owners home, as a nonce value: * Add $100 to person's central bank account * Over the next 30 days person buys: shoes locally, a pizza locally, and orders a book from Amazon. * On day 31 the central bank examines all transactions. By referencing each account money was transfered to the bank can lookup distance and type of business. * Remove the unspent amount, if any, from the account. Now, there is a fun edge case here: what if the person spends their $100 credit on books from Amazon instead of food and has no money left? Do they end up with a negative balance? This penalizes both people mispending money as well as those where they spent money at a place that is miscategorized ( for example, the local deli is categorized as a shell corporation ). So clearly the basic idea needs some work, but it is not undoable. Especially if business owners are incentived to have their information be accurate and up to date to recieve monetary funds.