Rev. Proc. 2008-50 provides a useful Appendix B , Section 3 to help us calculate earnings in the case of corrections. However, that section does not refer to losses, which must be taken into account for some, but not all corrections. Do you think it is appropriate to simply "read into" the language that any losses are to be calculated the same way. For example, if you choose the method that calls for using highest investment return among various funds, you would use the "least loss" where there is a negative return for all funds? On second thought, it is unlikely that the money market fund would actually be in a negative position, so I suppose that fund would be the rate of return to use in calculating earnings/losses to be applied to the contribution. Does this make sense?

Makes sense to me. FWIW

Makes sense to me to -- which means it is likely wrong in IRS logic.

There are other threads discussing that the IRS has a stated rate of return to use (can't remember if it's published or what). Do a search on here for correction interest rate. There was thread that touched on it in the past two weeks if I remember correctly.