Deferring Capital Gains Taxes

                                  Portfolio A    Portfolio B
Beginning amount                     $8,000        $8,000
Gain in year 1 (assume 25%)          $2,000        $2,000
Total after year 1 (before taxes)   $10,000       $10,000
Realized gain in year 1                  $0        $2,000
Taxes paid (assume 40%)                  $0          $800
Net after year 1                    $10,000        $9,200
Gain in year 2 (assume 25%)          $2,500        $2,300
Total after year 2 (before taxes)   $12,500       $11,500
Realized gain (end of year 2)        $4,500*       $2,300
Taxes paid (assume 40%)**            $1,800          $920
Total taxes paid (year 1 + year 2)   $1,800        $1,720
Net amount at end of year 2         $10,700       $10,580

** This is a simplification. Portfolio A would likely have more long-term capital gains, which would reduce the taxes significantly since the long-term capital gains tax rate is only 20%.

Having $120 (1.1%) more might not sound like much, but add in the power of compounding over decades -- not to mention the benefit of paying the much lower tax rate on long-term capital gains, an added benefit not reflected in the scenarios above -- and the difference is enormous. Consider three scenarios whereby you earn 10% returns per year for the next 20 years on an initial $1,000 investment. If your profits were taxed annually as short-term capital gains (39.6%), in 20 years your investment would be worth $3,231. If it were taxed annually as long-term capital gains (20% rate), it would be worth $4,661, 44% more. And if you were to instead buy one stock that compounded at 10% annually, and then after 20 years sold it and paid the 20% capital gains tax, you would have $5,582, 73% more.

OK, so now let's say you're persuaded -- and perhaps regretting incurring some of those gains earlier this year. What can you do to reduce your tax burden? One answer is to sell stock that has declined -- none of that in my portfolio, of course -- and then deduct the loss from the gains. But, if you're anything like me, you may be bullish on some of your most beaten-down stocks and therefore not be interested in selling them. So what can you do?

You can sell a stock and buy it back in 30 days (if you buy it back any sooner, you can't take the capital loss). But there's a major risk with this strategy: The stock could jump during those 30 days, and you won't benefit from it. That's the worst of both worlds: incurring the loss on the way down but not getting the gain on the way up. Is there a way to mitigate this risk?

I can think of one way, which involves selling a stock for a loss, buying a close substitute, and then swapping back in 30 days. For example, Fannie Mae (NYSE: FNM) and Freddie Mac (NYSE: FRE) are in the same business, share similar risk profiles, are trading at comparable valuations, and their stocks have tended to track each other pretty closely. In fact, over the past eight years, including this year to date, the difference in the annual stock performance between the two has only once exceeded 13 percentage points (last year, when Freddie Mac was up 51% vs. Fannie Mae's 27%). Thus, if you really like one of these stocks and want to hold it, but you have capital losses (through last Friday, Freddie Mac was down 25% and Fannie Mae was off 13% from their highs earlier this year), you could sell one, buy the other, and then swap back in 30 days.

What are some other example of possible substitutes? I haven't done extensive research on this, but if you have a loss on a large-cap technology stock (hard to imagine, isn't it?), you could trade for some shares of the Nasdaq 100 Trust (AMEX: QQQ). Or, if you hold Merck, you could trade for Bristol-Myers Squibb or one of the other large pharmaceutical companies. You get the idea.

This double-swap strategy is not without risks, however. First, no stock is a perfect substitute for another. Even the stocks of Fannie Mae and Freddie Mac -- two of the closest substitutes I have ever seen -- do not always track one another. For example, in October, Fannie Mae rose 12.8% while Freddie Mac rose only 4%. Second, there are trading costs -- both commissions and the bid/ask spread -- which might total a few percentage points depending on which stocks you're swapping, the size of your trades, and how much your broker charges. Finally, if the substitute stock rises, you will incur short-term capital gains taxes when you switch back to your original stock.

But you can mitigate most of these risks with a variation that I call the swap-and-hold strategy, which involves buying a substitute stock that you intend to hold because you like it as much or better than your current stock. Or, if you like them equally and want to own both stocks, use new cash to buy back the original stock once 30 days have elapsed. For example, let's say you own Freddie Mac and have a loss on it. You could sell it, buy an equal amount of Fannie Mae, and hold it. Then, if you wish, use other cash to buy Freddie Mac anytime after 30 days.

If you are thinking about just doing a double-swap, here are the criteria I would apply to decide whether to do so:

1. You have a high cost basis on the shares you might sell, and can therefore generate significant capital losses by selling;
   
   
2. You are in a high tax bracket and are generating short-term losses to offset short-term gains (in my opinion, it's not worth doing this to avoid paying the relatively low 20% long-term capital gains tax rate);
   
   
3. You are selling all or most of your current position (if you are only selling a small fraction of it, then you will already be getting most of the benefit if the stock jumps);
   
   
4. You can find a good substitute stock to buy, but have a strong preference for holding only your current stock, not both; and
   
   
5. You have a large base of capital, such that the trading costs are insignificant.