The simple version of the theory is that stock buybacks should be price neutral because shareholders effectively lose cash now (for the buyback) but get a higher proportion of future earnings for that cash. If the price of the stock is perfectly efficient, these values should be equal.