step up cost basis of stock

I've understood it to be the average of the high/low for the date in question, although I cannot find a citation for this.

I am not aware of any specific IRS guidance that calls for using the average of the high and low for the trading day. That number represents a reasonable estimate of the fair market value (FMV) and has not been questioned by the IRS. I would argue that it is not the only method for arriving at the FMV.

E.g., there are many stocks that have very small amounts of trading on any given day and the swing between high and low can be dramatic. This type of stock looks more like closely held stock and any method that the IRS has accepted for closely held stock would also fit the bill. Another example would be a stock that opened way down ( its low for the day) because of information that was released after the close of business the day before. Immediately after opening down, the stock recovers and continues to go up with most of the volume at a price that is considerable higher than the opening low. I would argue that using the median average would be a more reasonable representation of the FMV.

All this being said, using the average of high and low is easy because those numbers are readily available everywhere. Where the FMV of any security starts to become critical relative to gain or loss, one should look at the activity for the day to see whether a weighted average price or the median provides a better result.

I use the price at close, as that is the last trade price, and therefore the most accurate valuation until the market reopens. That's what brokers list on their statements as "current value."

See

Much thanks, Art. For me, it's more about closure and a citation. 706 is where I must have seen this.

The source of that instruction is Treas. Reg. 20-2031-2(b)(1), (b)(2) & (b)(3), Valuation of Stocks & Bonds. Don't know how I missed that....

However, quoting from 20-2031-2(e):

(e) Where selling prices or bid and asked prices do not reflect fair market value.

If it is established that the value of any bond or share of stock determined on the basis of selling or bid and asked prices as provided under paragraphs (b), (c), and (d) of this section does not reflect the fair market value thereof, then some reasonable modification of that basis or other relevant facts and elements of value are considered in determining the fair market value.

Right, and doesn't that effectively mean CLOSING value - the last time bid and ask succeeded in causing a transaction...?

Probably not. A federal court would have to give deference to the regulation. You would have to show that the last transaction was a much better representation of FMV for the date of death, than the mean average of the high and low. I don't think the last transaction of the day would meet that test. I can see where some other average (median or volume weighted) might be a candidate. And, as I said in my original reply, the method for closely held stock would probably be a better candidate for stock with thin volume.

Then how do you define "fair market value?"

I define it as a price upon which a willing seller can find a willing buyer, and the last trade constitutes the price where a willing seller FOUND a willing buyer, so why isn't the last price (and closing price when the market is closed) by definition the best example of FMV?

Probably because it is the FMV for the day the person died, not 3:59:59 PM eastern time.

By that definition, _every_ sale that day was at FMV. So why favor the last?

Seth

That time occurs once per day, even on the date of death.

True ("every sale" was FMV at the time it occurred).

Why favor the last? Because FMV changes. If it were constant, we wouldn't need rules to determine it.

That is not a way in which it differs from the time of any other trade.

Seth

Other than it's the last trade of the day.