There is no qualified dividend rate table, and understanding that explains several things at once. IRC 1(h)(11)(A) does not set rates for dividends. What it says is that, for purposes of section 1(h), which is the subsection containing the capital gains rate schedule, "the term 'net capital gain' means net capital gain (determined without regard to this paragraph) increased by qualified dividend income." Qualified dividends are poured into the same container as long-term capital gains and then run through the same brackets.
Two consequences follow, and neither is obvious from the popular phrasing that qualified dividends "get capital gains rates."
The first is that dividends and long-term gains share the rate bands rather than each receiving their own. Room at the lowest rate is consumed by the combined figure, so realizing a large long-term gain can push qualified dividends that would otherwise have been taxed at the lowest rate up into the next band, and vice versa. Anyone planning around the bottom of the capital gains schedule needs to count dividends they have no intention of doing anything about, because the statute has already counted them.
The second runs the other way and is a limit rather than a benefit. The merger at 1(h)(11)(A) is expressly "for purposes of this subsection," which is the rate subsection. It does not touch the netting rules in sections 1211 and 1222, which produce the net capital gain figure in the first place. So capital losses net against capital gains, and only the survivor of that netting is then increased by qualified dividend income. A capital loss therefore does not offset a qualified dividend the way it offsets a capital gain, even though the two are taxed at the same rates. Investors who think of the two as interchangeable get this backwards.
The holding period is borrowed rather than written, which is why it reads so oddly. IRC 1(h)(11)(B)(iii) does not state a holding period. It imports section 246(c), a provision written for the corporate dividends-received deduction, by substituting "60 days" for "45 days" each place it appears and "121-day period" for "91-day period." That is the source of the rule stated on the dividends page, and it also means the rest of section 246(c) comes along with it. In particular, section 246(c)(4) does not count days on which the holder's risk of loss was diminished, including periods when the holder had an option to sell or a contractual obligation to sell substantially identical stock, had an open short sale, or had granted an option to buy. So hedging a position can cost the qualified treatment on its dividends without any sale taking place, which is a consequence a reader would never derive from the 60-day rule alone.
A foreign payer has to get in through one of three doors. IRC 1(h)(11)(C) treats a foreign corporation as qualified if it is incorporated in a possession of the United States, or if it is eligible for the benefits of a comprehensive income tax treaty with the United States that the Secretary determines is satisfactory and that includes an exchange of information program, or if the dividend is paid on stock that is readily tradable on an established securities market in the United States. Two categories are excluded outright: a passive foreign investment company, for the year the dividend is paid or the preceding year, and a corporation that first becomes a surrogate foreign corporation after the enactment of that provision.
That test is the reason an international fund's qualified proportion can differ sharply from a domestic fund's, and can differ between two international funds. A holding whose shares trade on a US market, typically through a depositary receipt, gets in through the third door regardless of its home country's treaty position. A holding that trades only abroad depends on the treaty. And a fund with meaningful exposure to companies treated as passive foreign investment companies will report a smaller qualified share, because those dividends cannot qualify at all.
One election trades the rate away deliberately, and it catches investors who borrow to invest. IRC 1(h)(11)(D)(i) provides that qualified dividend income does not include any amount the taxpayer takes into account as investment income under section 163(d)(4)(B). The investment interest expense deduction is limited to net investment income, and qualified dividends are not automatically part of that figure. A taxpayer may elect to include them in order to deduct more investment interest, but the dividends brought in are then taxed at ordinary rates. It is a genuine either/or on the same dollars, and which side is better depends on the gap between the two rates and the size of the interest being deducted.