It's still too early to determine if the program will be successful in the long term, but early results are that it's been a lot more effective a creating jobs than any other traditional jobs program.

This got me thinking: not only should someone (YC maybe?) try this with startups in the US, but companies should also do something similar with employees.

Using a random sample would allow companies to determine if their hiring practices are really effective at all or just a waste of time. It would also eliminate any biases from the hiring manager.

We know that stock picking isn't any more effictive than random chance (that's the theory behind index funds). Maybe the same is true for investing in startups and hiring employees.