Suppose you and your friends all want to start small businesses.  You've all got money coming in, you're all in a position to save up some capital, but it would take you a year to all save up enough--and you'd rather not wait.  There's a solution to this problem that's so obvious its been reinvented all over the world.

In Mexico it's called a tanda or a cundina.  In Korea it's a kye.  In China it's a hui.  In West Africa it's a susu.  Here's an example of how one works (although the details endlessly variable):

A dozen friends arrange to meet every month for a year.  At every meeting, each one contributes $500.  The pool of capital ($6000) goes to one member, who now has enough money to start his or her business.  The process is repeated each month for a year, until each person has gotten the money.

The person who gets the money is usually chosen at random from among the people who haven't gotten it yet, but that's not the only way to do it.  The members could determine the order based on who is most prepared to get their new business started.  Or do it alphabetically.  Basically, anything they can agree on will work.

The person who gets the money last is perhaps slightly less well off than he would have been just saving the money for a year himself--losing out on the bit of interest that would have been earned--but that's not going to amount to enough to affect the success or failure of the business.  (My financial calculator says that he'd have earned $83, if he'd put the money into a savings account paying 3%.)  Essentially, the last guy comes out even and everyone else comes out ahead.

The generic term for these schemes is a ROSCA--a Rotating Savings and Credit Association.  I first ran across this concept in the Korean version, when I was living in Los Angeles, then started hearing about it other places as well.  It seemed like magic to me:  Everybody comes out ahead!  It was so obviously a good idea, I didn't understand why it was only turning up in immigrant communities.  The closest American analog I could think of is a barn raising (where everybody donates a day's labor against a general community obligation to reciprocate).

After thinking about it, though, I realized that it's really an alternative to having access to banks.  

A scheme like this means that the average participant has to wait 6 months.  The obvious alternatives would be either to just save the entire sum (waiting a year) or borrow the money from a bank (not waiting at all--putting $500 down, borrowing $5500, and paying the loan off over the course of a year).  Assuming that the business is reasonably profitable, neither the interest earned during the year of saving nor the interest paid while paying off the loan would make a difference that amounted to success or failure.  Any of the three alternatives is about as good the other.  And, if all twelve borrow the money from the bank, nobody needs to wait.

If bank credit is unavailable, though, this sort of scheme really shines.

Similarly, if bank saving is unavailable (because the banks aren't safe, or they won't let you open an account because you don't have the right kind of ID), then this is again an attractive alternative to the saving option.  (It's also an effective measure for dealing with other kinds of obstacles to savings, such as family members who insist on immediately spending any money that you put aside.)

Another reason that these things only show up in some communities and not others is that it requires a considerable degree of trust--in the other participants, but also in the community as a whole to provide the appropriate social pressure.  (In some American subcultures, I think any social pressure would be exactly backwards--friends and neighbors would consider someone a chump to keep putting in $500 a month after he'd already gotten his $6000.  Even a whiff of that attitude would make a scheme like this impossible.)

I've written before about how sharing can make everyone better off.  This is a rather different example of the same principle.  Everybody can come out ahead.