Some good suggestions here, but I think the best were kind of scattered across a few posts - I'll give my 2 cents and try to prioritize/organize these previous suggestions for you.
First, you mention that he
may need to sell some/all to support himself, so you may not have the option of taking the step-up in cost basis. So first I would (nothing new here, I don't think):
1) Keep some perspective, worst case a default cost-basis of zero $, means worst case a 15% tax hit on any sale, which may only be a small portion of the total (right?). Sure, you want to avoid that, but it's not the end of the world either. Can any be done at a 0% tax hit? Balance the effort against potential $ saved.
2) Stop all dividend re-investments
now. Document that they were always DRIP prior to that date. That kills two birds with one stone - he has more income, so less likely to need to sell any (or at least not as much), and it simplifies calculations going forward.
Originally Posted by Running_Man View Post
And if he just stops the DRIP he would have $5,644 a year in dividends to spend
That's big. In just 5 years, he'd have an extra ~ $28,220 he can tap before selling anything.
3) You should be able to get the dividend records going back quite a few years, and if you don't have a record of the reinvested price, estimate it, and you can sell those specific shares (that are > 1 year old). These would be recent, so tracking shouldn't be hard, and gains are probably low. You just list the date acquired as 'various', all long term, and your total purchase and sale prices. Keep a record for yourself, but if it is reasonable, and the IRS doesn't question it, you are fine. If they do question, you have the documentation, no problem. I've used the 'various' several times in the past, never was questioned, as the values were reasonable.
I think I would start now, selling off the recent reinvestments, to stay in the 0% or 15% hit, which is easier when done a little at a time.
4) And if he does need to sell more than the ones that can be documented as reinvestements with known (or at least very close) prices, then consider just using zero basis, the tax hit on that portion may not be worth the effort.
-ERD50