I agree. It's certainly fine for google to see that statistic and do some due diligence to make sure their are not any issues in their hiring managers/process. But the answer is almost certainly in the available pool of people who apply. I'd definitely be interested to see the percentage of people who applied that are white vs black, male vs female. If those percentages are way different than their work force make-up, then maybe there's something more to look at. However you'd then have to investigate a large number of individual examples to see if qualifications and interview results show a valid reason for the difference. All-in-all the likelihood that it involves racist HR/hiring managers at google is extremely unlikely.