[bstadil]

The number 1 as in 1 percent is the least suspicious number, so Benford's law is unlikely to help.

[reitzensteinm]

You can still work with the rest of the digits in the percentage (eg 1.1% should be more likely than 1.9%), I imagine the problem isn't the numbers themselves but the lack of a decent sized sample. You only get 12 numbers for a year and that's plenty of time to lose your shirt to a con man who has your money!

[byrneseyeview]

If they average around 1%, that would mean as many .7s as .3s -- very suspicious!

[byrneseyeview]

Please disregard this comment, which was idiotic.