Theory on taking capital gains/losses

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Saving$
Posts: 1530
Joined: Sat Nov 05, 2011 8:33 pm

Theory on taking capital gains/losses

Post by Saving$ » Sun Mar 17, 2013 10:29 am

There have been numerous times I've seen those trying to change taxable accounts to a boglehead type portfolio be advised to balance the sale of losers and winners to lessen the tax bite. So if you sold funds which increased in value by $3,000 and at the same time sold fund that lost $3,000 in value, you would have no tax impact.

However, it looks to me there might be a better strategy if the timing works, and if you don't have other sales for the years in question. Assume you are in the 25% tax bracket:
a. In 2013, sell funds which have increased in value by $3,0000. This $3,000 is taxed as a capital gain at 15%, costing you $450 in taxes.
b. In 2014, sell funds which have decreased in value by $3,000. Now you have a capital loss, which seems to be an offset in income. So if you are in the 25% bracket, you just saved $750 on your taxes.
In this example, the net savings of splitting the sales among two tax years would be $300. If your losses were greater, you could roll the capital loss over, and have more savings.

Am I missing something?

STC
Posts: 415
Joined: Wed Nov 14, 2012 9:22 am

Re: Theory on taking capital gains/losses

Post by STC » Sun Mar 17, 2013 11:05 am

That "if the timing works out" is a biggie.

1: anyone carrying forward losses cannot do this
2: to produce both a $3k gain and a $3k loss you would likely have to own investments that are volatile and negatively correlated - like extended duration bonds and equities - or have a lot of money invested in low volatility and negatively correlated assets - like intermediate muni's and equities.

Wagnerjb
Posts: 7027
Joined: Mon Feb 19, 2007 8:44 pm
Location: Houston, Texas

Re: Theory on taking capital gains/losses

Post by Wagnerjb » Sun Mar 17, 2013 11:21 am

You got it exactly right. It is tax arbitrage...paying at one rate, saving at a higher rate. This - along with the time value of money - is the fundamental principle behind tax loss harvesting.

Best wishes.
Andy

User avatar
Peter Foley
Posts: 4101
Joined: Fri Nov 23, 2007 10:34 am
Location: Lake Wobegon

Re: Theory on taking capital gains/losses

Post by Peter Foley » Sun Mar 17, 2013 11:35 am

Saving$ wrote:
However, it looks to me there might be a better strategy if the timing works, and if you don't have other sales for the years in question. Assume you are in the 25% tax bracket:
a. In 2013, sell funds which have increased in value by $3,0000. This $3,000 is taxed as a capital gain at 15%, costing you $450 in taxes.
b. In 2014, sell funds which have decreased in value by $3,000. Now you have a capital loss, which seems to be an offset in income. So if you are in the 25% bracket, you just saved $750 on your taxes.
In this example, the net savings of splitting the sales among two tax years would be $300. If your losses were greater, you could roll the capital loss over, and have more savings.

Am I missing something?


I agree with your analysis and have done this in reverse order in the past. During the market downturns in 2000 and again in 2009 I did tax loss harvesting. The losses were great enough that it took me a few years to use them up. During those time periods I avoided taking capital gains in taxable accounts. I think it is harder to do in the order you describe in a. and b. above.

livesoft
Posts: 57297
Joined: Thu Mar 01, 2007 8:00 pm

Re: Theory on taking capital gains/losses

Post by livesoft » Sun Mar 17, 2013 11:59 am

Yep, try to do this in reverse order by selling losers first. :)
This signature message sponsored by sscritic: Learn to fish.

Saving$
Posts: 1530
Joined: Sat Nov 05, 2011 8:33 pm

Re: Theory on taking capital gains/losses

Post by Saving$ » Sun Mar 17, 2013 12:28 pm

I agree with selling the losers before the winners as long as you are SURE that the losers will not generate more than $3k loss. If there is any chance of more than $3k loss, sell the winners first, because there is no limit on the 15% cap gains. The second year, you sell the losers, and if the loss is greater than $3k, you keep carrying it over until used. But if the point is to get to a Boglehead portfolio, you get it done in the second year.

If you sell the losers first and have more than $3k loss, you delay getting to a Boglehead portfolio until the year after you use up all your losses, OR you leave money on the table with the IRS. If the winners remain winners, the delay is no big deal, but you have no way of knowing that.

User avatar
grabiner
Advisory Board
Posts: 20999
Joined: Tue Feb 20, 2007 11:58 pm
Location: Columbia, MD

Re: Theory on taking capital gains/losses

Post by grabiner » Sun Mar 17, 2013 6:02 pm

The problem with splitting the timing is that you don't know what your gains will be next year. If you have $3000 of losses in Fund A and $3000 of gains in Fund B, and you sell them both now, you will have a zero tax bill. If you sell only Fund A and wait until next year to sell Fund B, you will save $300, but only if Fund B doesn't rise in value between now and next year. Usually, you expect the market to rise, and if Fund B rises in value by $3000 before you sell it, you'll pay an extra $450 in taxes, more than negating the $300 gain. Meanwhile, if you sell Fund B today and the market falls next year, that doesn't cost you anything because you can harvest the loss in the replacement fund.
David Grabiner

Post Reply