Message Font: Serif | Sans-Serif

No. of Recommendations: 0
Luckily the tax law passed earlier today is much saner than some of the initial proposals.

One big problem I perceived to exist in the initial proposal of full exclusion of dividends from tax is the apparently ease of converting short-term capital gains into tax-free dividends. Please correct me if I am wrong, but the following strategy appears to have been valid using that version of proposed tax law.

1) Have short-term capital gains in issue X.
2) Sell X and realize those short-term gains.
3) Buy Y just before it declares a dividend ("buy a dividend").
5) Sell Y at a short-term capital loss.
6) Balance short-term capital gains from X with the short-term capital loss from Y. No tax due.
7) Dividend from Y. No tax due.

Is there some sort of fallacy here I missed ?
No. of Recommendations: 0
i think you are right theoretically that is a valid strategy.

in the process you do expose yourself to some market risk and the transaction costs. sound theoretically, but tricky to implement. neat idea though.
No. of Recommendations: 0
Luckily the tax law passed earlier today is much saner than some of the initial proposals.

One big problem I perceived to exist in the initial proposal of full exclusion of dividends from tax is the apparently ease of converting short-term capital gains into tax-free dividends. Please correct me if I am wrong, but the following strategy appears to have been valid using that version of proposed tax law.

1) Have short-term capital gains in issue X.
2) Sell X and realize those short-term gains.
3) Buy Y just before it declares a dividend ("buy a dividend").
5) Sell Y at a short-term capital loss.
6) Balance short-term capital gains from X with the short-term capital loss from Y. No tax due.
7) Dividend from Y. No tax due.

Is there some sort of fallacy here I missed ?

No fallacy, the problem is in the execution. Since most dividend yields are around 1%/year, but are paid quarterly, you would probably have to invest much more capital in Y to generate a dividend (and corresponding STCL) equivalent to the STCG realized in X.

Ira
No. of Recommendations: 2
Since most dividend yields are around 1%/year, but are paid quarterly, you would probably have to invest much more capital in Y to generate a dividend (and corresponding STCL) equivalent to the STCG realized in X.

There are many dividends much better than 1%: each of Altria (formerly Philip Morris), General Motors, Kodak, AT&T, SBC, and JP Morgan Chase are currently paying anywhere from 4% to 7%. But Ira's right - buying a dividend large enough to cover a STCG would require a large investment. You could ease that a bit by buying on margin, but that carries its own risks. Moreover, in recent years the volatility of the market (and the generally lower yields) makes this strategy difficult. The price of a stock should (theoretically) drop by the amount of the dividend when the stock goes ex div - but often normal day to day variations in price far exceed that amount.

Lorenzo
No. of Recommendations: 0
The price of a stock should (theoretically) drop by the amount of the dividend when the stock goes ex div - but often normal day to day variations in price far exceed that amount.

Wouldn't you expect an appropriate price drop on the date of record?

Bill
No. of Recommendations: 1
Wouldn't you expect an appropriate price drop on the date of record?

No, just on the ex div date. Example:

Altria Group (MO, formerly Philip Morris) paid a dividend of 64 cents/share on April 9. The record date for that dividend was March 14, which meant that the ex dividend date was March 12. (Ex div date is typically two business days before record date.) Sure enough, MO closed at \$36.10 on March 11, and opened at \$35.49 on March 12, a drop of 61 cents (roughly equal to the dividend). That's because if you bought MO on March 11 (before the ex div date), you would have received the 64 cent dividend on April 9. If you bought on March 12 (the ex div date), you wouldn't get that dividend, and so it's sensible that you would pay less.

The reason for the two day difference between ex div and record dates is that it takes three days to settle. If you bought MO on Mar 11, the trade would have settled on Mar 14. That's the record date - in other words, you own the stock at close of business and thus get the dividend...

MO pays one of the larger dividends among the Dow stocks, so the effect is pretty clear. The point of my earlier post was that a stock like AT&T (dividend 18.875 cents) often has day-to-day gyrations well in excess of its dividend amount. Indeed, T is up 74 cents today, the amount of its annual dividend.

Lorenzo
No. of Recommendations: 0
No fallacy, the problem is in the execution. Since most dividend yields are around 1%/year, but are paid quarterly, you would probably have to invest much more capital in Y to generate a dividend (and corresponding STCL) equivalent to the STCG realized in X.

First of all, any equity that pays a dividend (MO for example pays 6% or so) could be used to generate the dividend income. Second of all, the capital need only be deployed for one day or so when the dividend is recorded. And third of all, limited capital could be used "serially" to gain dividends (and capital losses) from multiple equities.
No. of Recommendations: 1
Luckily the tax law passed earlier today is much saner than some of the initial proposals.

One big problem I perceived to exist in the initial proposal of full exclusion of dividends from tax is the apparently ease of converting short-term capital gains into tax-free dividends. Please correct me if I am wrong, but the following strategy appears to have been valid using that version of proposed tax law.

1) Have short-term capital gains in issue X.
2) Sell X and realize those short-term gains.
3) Buy Y just before it declares a dividend ("buy a dividend").
5) Sell Y at a short-term capital loss.
6) Balance short-term capital gains from X with the short-term capital loss from Y. No tax due.
7) Dividend from Y. No tax due.

Is there some sort of fallacy here I missed ?

An article about the new tax law in today's NY Times indicates that you did miss something, or that the drafters had this dodge in mind. In order to get the reduced rate on dividends, you must hold the underlying stock for 60 days. I wonder how they're going to track all of this.

Ira

No. of Recommendations: 0
An article about the new tax law in today's NY Times indicates that you did miss something, or that the drafters had this dodge in mind. In order to get the reduced rate on dividends, you must hold the underlying stock for 60 days. I wonder how they're going to track all of this.