SCENARIO: Brokerage account was JTWROS for husband/wife. NOT a community property state. I was recently reviewing the calculated cost basis numbers, which in simplest terms keep half of the value of the stock unchanged and change the other half (up/down) based upon the market value on date of death of spouse.

METHOD I HAVE ALWAYS USED: For each purchase lot, keep one-half of the shares and the original (perhaps adjusted due to corporate actions) share cost unchanged. For the other half, the per share cost is determined by the average of the high/low price on the date of death. Those shares show a "purchase" date same as the date of death, and are deemed to be LTCG (even if the underlying purchase remains short-term for the "original" half.)

However, the gain/loss "unrealized" calculations, prepared by the brokerage firm, I have now show. . .different numbers.

It took a while, but I finally got an explanation as to how the brokerage firm had calculated the cost bases for the formerly JTWROS account now with single (widow) ownership.

"I have the answer for you. The way a 50% step-up was done at ----- is as follows: They took 50% of the original cost basis plus 50% of the stepped-up cost basis (determined using the average of the high and low prices on the date of death). They then took the average of those two values and applied the resulting cost to both tax lots. (That is why both lots for each security originally had the same share price, before we made adjustments on ------.) Then ------ recorded one lot as having the original holding period and the other as having the inherited date and status."

QUESTION #1 - Has anyone used this alternative method, and does the IRS sanction its use? (cite?)
QUESTION #2 - Since I now have a "repaired" (brokerage firm #2) version of an original (brokerage firm #1) set of cost bases numbers, I think something now needs to be changed. It may be possible for me to revert to the numbers calculated/shown by brokerage firm #1, which were transferred to brokerage firm #2. (I doubt if brokerage firm #2 has access to historical records of the now-defunct brokerage firm #1.)
QUESTION #3 - (For the number-crunchers out there with considerably more skills than myself!) Do the two valuation methods (traditional vs alternative) essentially generate the same overall [total] cost basis?

Any insight or suggestions as to what I should do next will be greatly appreciated!