tax-loss harvesting question
Moderator: Global Moderator
tax-loss harvesting question
I have some Treasuries I bought 5 years ago in a taxable account, so they are 25 years out right now. I've taken a big hit on the value of the Treasuries. Should I sell them now and immediately replace them with 30 year Treasuries so I can lock in the long term loss and use that loss to offset future long term gains? Due to the bid/ask spread, swapping out Treasuries is not something one wants to do frequently. So am I engaging in market timing by swapping them out now when they could theoretically go even lower? Did Harry Browne talk about tax-loss harvesting, generally, and in regards to Treasuries, specifically? I'm not sure what to do.
Re: tax-loss harvesting question
tax loss harvesting is something that you have to decide for yourself, I did it in a taxable account and replaced it with a close to identical holding so I still have my position in long term bonds. Sure they could go lower and in that case I plan on doing another tax loss harvest by selling what I bought (VGLT) and re-buying TLT if it doesn't go lower I will decide whether or not to keep VGLT or go back to TLT, I will re evaluate after 30 days. In the meantime I locked in a big loss which will help reduce my capital gains. I'm not a market timing just taking advantage of current situation. It is entirely up to you and what you understand and feel comfortable doing.
Re: tax-loss harvesting question
Sounds like a great idea, Jason. As long as 25- and 30-year treasuries aren't considered "substantially identical" for the wash sale rule -- and I don't see why they would be. If you're worried about the value of the bonds going lower still, maybe just sell half now and the rest next year or later this year?
Regarding bid/ask spreads, I have found them to be reasonable when trading LTTs this year at Fidelity, especially when trading larger lots. In the past I recall the spread was wide and I was reluctant to use the secondary market. Now I think it's not much different from trading stocks.
Regarding bid/ask spreads, I have found them to be reasonable when trading LTTs this year at Fidelity, especially when trading larger lots. In the past I recall the spread was wide and I was reluctant to use the secondary market. Now I think it's not much different from trading stocks.
-
mathjak107
- Executive Member
- Posts: 4456
- Joined: Fri Jun 19, 2015 2:54 am
- Location: bayside queens ny
- Contact:
Re: tax-loss harvesting question
remember , tax loss harvesting kicks the tax can down the road
it bumps up the cost basis on what you buy to replace what you sold so it may or may not be a good thing .
i kicked the tax can down the road when we sold our commercial lease rights we held .
i kicked it right into the capital gains bracket going from 15% to 23.80% on our income with the gain .
the real good deal is tax gain harvesting.. that is where you have room to take tax free gains in the zero capital gains bracket
it bumps up the cost basis on what you buy to replace what you sold so it may or may not be a good thing .
i kicked the tax can down the road when we sold our commercial lease rights we held .
i kicked it right into the capital gains bracket going from 15% to 23.80% on our income with the gain .
the real good deal is tax gain harvesting.. that is where you have room to take tax free gains in the zero capital gains bracket
Re: tax-loss harvesting question
Three things/issues to be aware of.
One, why replace 25-year Treasuries with 30-year? I presume it is because you would rather buy brand new 30-year Treasuries at issue rather than buy (in the secondary market since the U.S. doesn't issue new 25-year maturity Treasuries) a 30-year Treasury that has aged five years and now has only 25 years left until maturity? IIRC Harry Browne said to hold them until they had 20 years left and only then sell to be replaced by new 30-year bonds.
Two, if you would in fact be willing to buy an aged Treasury bond with 25 (or 26, or 27, etc) years left, be aware of the possible positive or negative tax implications (vs just holding the current bonds you already hold) of doing so; these will depend on:
A. Your state tax bracket,
B. Whether you buy a Treasury at a discount to current market yield (i.e. buying a Treasury due in 2050 that was originally issued at a coupon of 1.25% in 2020 and that is thus priced hugely below par) vs one closer to current market yields (say the one just issued back in August and due in 2053) vs the most recent issue that will be available in November (theoretically there could be further tax implications still if Treasury bonds were available at a premium but given current rates no long Treasury issued in the last fifteen years is trading at a premium as of right now),
C. Whether you will be deducting the income from the tax loss sale against qualified long term capital gains, short-term capital gains, or (at a $3K per year limit) ordinary income,
D. What your tax bracket is now vs what you reasonably expect it to be a few years from now
Three, while theoretically there should be no reported wash sale (i.e. the brokerage won't report it directly to the IRS) if you buy a Treasury bond with just a year or two's difference in maturity from your currently owned ones you will be selling for a tax loss (since the CUSIPs are different.....brokerages don't automatically report wash sales unless the exact same security--with the exact same CUSIP--is re-bought), the IRS could (if you ever get audited) technically try to claim it was a wash sale using the precedent set in Hanlin v. Commissioner of Internal Revenue (1938); see https://www.anylaw.com/case/hanlin-v-co ... TomsSBhwbQ and https://casetext.com/case/hanlin-v-comm ... al-revenue . Unlikely, but still theoretically possible. In theory, the precedent doesn't directly apply (since the bonds at issue in the Hanlin case were municipal bonds and Federal instrumentality bonds--i.e. agency debt similar to GNMAs--rather than actual U. S. Treasury securities) but the IRS might try to argue in court that it does apply and then you'd have to decide whether it would be worth your time and money to challenge it in Tax Court. AFAIK they (the Internal Revenue Service) have not ever done this yet and in any event the brokerage won't report it outright unless it is a true CUSIP-for-CUSIP wash sale but this is just something to possibly be aware of.
One, why replace 25-year Treasuries with 30-year? I presume it is because you would rather buy brand new 30-year Treasuries at issue rather than buy (in the secondary market since the U.S. doesn't issue new 25-year maturity Treasuries) a 30-year Treasury that has aged five years and now has only 25 years left until maturity? IIRC Harry Browne said to hold them until they had 20 years left and only then sell to be replaced by new 30-year bonds.
Two, if you would in fact be willing to buy an aged Treasury bond with 25 (or 26, or 27, etc) years left, be aware of the possible positive or negative tax implications (vs just holding the current bonds you already hold) of doing so; these will depend on:
A. Your state tax bracket,
B. Whether you buy a Treasury at a discount to current market yield (i.e. buying a Treasury due in 2050 that was originally issued at a coupon of 1.25% in 2020 and that is thus priced hugely below par) vs one closer to current market yields (say the one just issued back in August and due in 2053) vs the most recent issue that will be available in November (theoretically there could be further tax implications still if Treasury bonds were available at a premium but given current rates no long Treasury issued in the last fifteen years is trading at a premium as of right now),
C. Whether you will be deducting the income from the tax loss sale against qualified long term capital gains, short-term capital gains, or (at a $3K per year limit) ordinary income,
D. What your tax bracket is now vs what you reasonably expect it to be a few years from now
Three, while theoretically there should be no reported wash sale (i.e. the brokerage won't report it directly to the IRS) if you buy a Treasury bond with just a year or two's difference in maturity from your currently owned ones you will be selling for a tax loss (since the CUSIPs are different.....brokerages don't automatically report wash sales unless the exact same security--with the exact same CUSIP--is re-bought), the IRS could (if you ever get audited) technically try to claim it was a wash sale using the precedent set in Hanlin v. Commissioner of Internal Revenue (1938); see https://www.anylaw.com/case/hanlin-v-co ... TomsSBhwbQ and https://casetext.com/case/hanlin-v-comm ... al-revenue . Unlikely, but still theoretically possible. In theory, the precedent doesn't directly apply (since the bonds at issue in the Hanlin case were municipal bonds and Federal instrumentality bonds--i.e. agency debt similar to GNMAs--rather than actual U. S. Treasury securities) but the IRS might try to argue in court that it does apply and then you'd have to decide whether it would be worth your time and money to challenge it in Tax Court. AFAIK they (the Internal Revenue Service) have not ever done this yet and in any event the brokerage won't report it outright unless it is a true CUSIP-for-CUSIP wash sale but this is just something to possibly be aware of.
-
vnatale
- Executive Member
- Posts: 9491
- Joined: Fri Apr 12, 2019 8:56 pm
- Location: Massachusetts
- Contact:
Re: tax-loss harvesting question
D1984 wrote: ↑Tue Oct 10, 2023 7:09 am
Three, while theoretically there should be no reported wash sale (i.e. the brokerage won't report it directly to the IRS) if you buy a Treasury bond with just a year or two's difference in maturity from your currently owned ones you will be selling for a tax loss (since the CUSIPs are different.....brokerages don't automatically report wash sales unless the exact same security--with the exact same CUSIP--is re-bought), the IRS could (if you ever get audited) technically try to claim it was a wash sale using the precedent set in Hanlin v. Commissioner of Internal Revenue (1938); see https://www.anylaw.com/case/hanlin-v-co ... TomsSBhwbQ and https://casetext.com/case/hanlin-v-comm ... al-revenue . Unlikely, but still theoretically possible. In theory, the precedent doesn't directly apply (since the bonds at issue in the Hanlin case were municipal bonds and Federal instrumentality bonds--i.e. agency debt similar to GNMAs--rather than actual U. S. Treasury securities) but the IRS might try to argue in court that it does apply and then you'd have to decide whether it would be worth your time and money to challenge it in Tax Court. AFAIK they (the Internal Revenue Service) have not ever done this yet and in any event the brokerage won't report it outright unless it is a true CUSIP-for-CUSIP wash sale but this is just something to possibly be aware of.
Quite the comprehensive analysis for what you yourself say is a fairly much low probability event!
Above provided by: Vinny, who always says: "I only regret that I have but one lap to give to my cats." AND "I'm a more-is-more person."
-
mathjak107
- Executive Member
- Posts: 4456
- Joined: Fri Jun 19, 2015 2:54 am
- Location: bayside queens ny
- Contact:
Re: tax-loss harvesting question
what looks weird is tlt shows 100% AA since the downgrade , yet vglt shows 99 %. AAA
Re: tax-loss harvesting question
I had looked that info up the last time I tax loss harvested long bonds. As long as the years are different (e.g. sell a bond maturing in 25 years and replacing with a bond maturing at 30 years) those are considered different investments. If you're selling a bond FUND though, and intend to buy bonds to replace it, you either have to buy an individual bond (obviously different from the fund) or use a different fund with a different underlying index.
The best reason to tax loss harvest is that you get to offset not only capital gains, but also ordinary income up to the $3K limit per year. If you can only offset capital gains, that may not be much of a win unless you expect to have some 0%-bracket capital gains in future. The way it works is that the tax loss has to be applied first to capital gains, then when you run out of those, you can offset ordinary income. So the answer also depends on how much cap gains you have.
Thanks for the reminder - I have a long bond sitting in my taxable investment account like a brick, and I'm thrilled to be able to get rid of it. My plan as been to use a tax-deferred account (a Keogh) to house long bonds, because if you end up losing money on those it's actually better than a tax loss harvest situation.
Let me explain that: Let's say you are retirement age, and you put $10,000 in stocks in a Roth IRA, and $10,000 in long bonds in the Keogh. The way the PP works is that losses in one category counterbalances losses in another. If the stocks go up to $15,000 in the Roth and you decide to cash it all out, you won't pay a dime in taxes. At the same time, let's say that the long bonds dropped in value to $5,000. You cash out the long bond and withdraw the money. That $5000 will now be subject to ordinary income tax rates. So you've ended up with the same amount of money you put in, but now you've avoided paying ordinary income tax (let's say 22%) on a quarter of the funds, instead of having paid ordinary income tax on the entire $20K.
Of course, it could work the other way round if instead the stocks go down and the bonds go up. But, that scenario is less likely. And, you can protect yourself from realizing those losses in the Roth/gains in the Keogh by spending from the cash allocation instead.
The best reason to tax loss harvest is that you get to offset not only capital gains, but also ordinary income up to the $3K limit per year. If you can only offset capital gains, that may not be much of a win unless you expect to have some 0%-bracket capital gains in future. The way it works is that the tax loss has to be applied first to capital gains, then when you run out of those, you can offset ordinary income. So the answer also depends on how much cap gains you have.
Thanks for the reminder - I have a long bond sitting in my taxable investment account like a brick, and I'm thrilled to be able to get rid of it. My plan as been to use a tax-deferred account (a Keogh) to house long bonds, because if you end up losing money on those it's actually better than a tax loss harvest situation.
Let me explain that: Let's say you are retirement age, and you put $10,000 in stocks in a Roth IRA, and $10,000 in long bonds in the Keogh. The way the PP works is that losses in one category counterbalances losses in another. If the stocks go up to $15,000 in the Roth and you decide to cash it all out, you won't pay a dime in taxes. At the same time, let's say that the long bonds dropped in value to $5,000. You cash out the long bond and withdraw the money. That $5000 will now be subject to ordinary income tax rates. So you've ended up with the same amount of money you put in, but now you've avoided paying ordinary income tax (let's say 22%) on a quarter of the funds, instead of having paid ordinary income tax on the entire $20K.
Of course, it could work the other way round if instead the stocks go down and the bonds go up. But, that scenario is less likely. And, you can protect yourself from realizing those losses in the Roth/gains in the Keogh by spending from the cash allocation instead.
Re: tax-loss harvesting question
Wow, Sophie!
Thank you for that thought and explanation.
It's brilliant and I had never thought of it that way previously.
Thank you for that thought and explanation.
It's brilliant and I had never thought of it that way previously.
Re: tax-loss harvesting question
The reason why I wanted to sell the 25 year Treasuries and buy 30 year Treasuries is so I won't have to sell and buy bonds again in 5 years, which is what I would have to do if I replaced them with different 25 year Treasuries (wash sale considerations aside). Every time you sell or buy bonds, there is a cost involved due to the spread. So why not just minimize the frequency of selling/buying when doing tax-loss harvesting with bonds by always replacing with 30 year bonds? Is there a reason not to?D1984 wrote: ↑Tue Oct 10, 2023 7:09 am Three things/issues to be aware of.
One, why replace 25-year Treasuries with 30-year? I presume it is because you would rather buy brand new 30-year Treasuries at issue rather than buy (in the secondary market since the U.S. doesn't issue new 25-year maturity Treasuries) a 30-year Treasury that has aged five years and now has only 25 years left until maturity? IIRC Harry Browne said to hold them until they had 20 years left and only then sell to be replaced by new 30-year bonds.
Two, if you would in fact be willing to buy an aged Treasury bond with 25 (or 26, or 27, etc) years left, be aware of the possible positive or negative tax implications (vs just holding the current bonds you already hold) of doing so; these will depend on:
A. Your state tax bracket,
B. Whether you buy a Treasury at a discount to current market yield (i.e. buying a Treasury due in 2050 that was originally issued at a coupon of 1.25% in 2020 and that is thus priced hugely below par) vs one closer to current market yields (say the one just issued back in August and due in 2053) vs the most recent issue that will be available in November (theoretically there could be further tax implications still if Treasury bonds were available at a premium but given current rates no long Treasury issued in the last fifteen years is trading at a premium as of right now),
C. Whether you will be deducting the income from the tax loss sale against qualified long term capital gains, short-term capital gains, or (at a $3K per year limit) ordinary income,
D. What your tax bracket is now vs what you reasonably expect it to be a few years from now
Three, while theoretically there should be no reported wash sale (i.e. the brokerage won't report it directly to the IRS) if you buy a Treasury bond with just a year or two's difference in maturity from your currently owned ones you will be selling for a tax loss (since the CUSIPs are different.....brokerages don't automatically report wash sales unless the exact same security--with the exact same CUSIP--is re-bought), the IRS could (if you ever get audited) technically try to claim it was a wash sale using the precedent set in Hanlin v. Commissioner of Internal Revenue (1938); see https://www.anylaw.com/case/hanlin-v-co ... TomsSBhwbQ and https://casetext.com/case/hanlin-v-comm ... al-revenue . Unlikely, but still theoretically possible. In theory, the precedent doesn't directly apply (since the bonds at issue in the Hanlin case were municipal bonds and Federal instrumentality bonds--i.e. agency debt similar to GNMAs--rather than actual U. S. Treasury securities) but the IRS might try to argue in court that it does apply and then you'd have to decide whether it would be worth your time and money to challenge it in Tax Court. AFAIK they (the Internal Revenue Service) have not ever done this yet and in any event the brokerage won't report it outright unless it is a true CUSIP-for-CUSIP wash sale but this is just something to possibly be aware of.
Re: tax-loss harvesting question
Not sure I understand your question.
Yes there is a bid-ask spread when you buy individual bonds, but if you do it infrequently enough, that cost is less than the expenses you would pay for owning shares in a bond fund.
If you're selling bonds to tax loss harvest you will either want to immediately put the money back into either an individual 30 year bond or a long Treasury fund, right? I'd just do whatever is convenient. The different in expenses isn't enough to get all tied up in knots about.
One word of caution: if you are buying long bonds at Fidelity, make sure you are buying at least $10,000 at a time. Fidelity's software makes it super difficult to unload bonds if you have less than that. I once bought a single bond ($1000 face value) and it took me months and several dozen tries before it finally sold (or I could have called them and spent the requisite maddening 2 hours on the phone during business hours, which is a near impossibility for me). Don't know if the other brokerages have the same issue. I guess there's only one way to find out.
Yes there is a bid-ask spread when you buy individual bonds, but if you do it infrequently enough, that cost is less than the expenses you would pay for owning shares in a bond fund.
If you're selling bonds to tax loss harvest you will either want to immediately put the money back into either an individual 30 year bond or a long Treasury fund, right? I'd just do whatever is convenient. The different in expenses isn't enough to get all tied up in knots about.
One word of caution: if you are buying long bonds at Fidelity, make sure you are buying at least $10,000 at a time. Fidelity's software makes it super difficult to unload bonds if you have less than that. I once bought a single bond ($1000 face value) and it took me months and several dozen tries before it finally sold (or I could have called them and spent the requisite maddening 2 hours on the phone during business hours, which is a near impossibility for me). Don't know if the other brokerages have the same issue. I guess there's only one way to find out.
Re: tax-loss harvesting question
I added to my bond holdings yesterday at Fidelity. The bid-ask spreads were very narrow and stock-like. I don't think we need to worry anymore about minimizing trading in the secondary bond market to avoid high spreads -- at Fidelity at least.
Also, the Fidelity bond trading platform has been modernized and the problems Sophie mentioned are almost completely eliminated. You can now trade a single bond as easily as 10 or more, knowing what the price will be. The only drawback is that you have to be quick with placing the trade: refresh the page to get the most recent bids/asks, select the bond based on minimum trade (there's a list of maybe six different lots available for trading, down to a single bond), enter the number of bonds to be traded, and go! But, in my experience, it must all be done within about 10 seconds, otherwise an error message might appear saying that the prices have changed. It's not too difficult to go back and make the selections again, just be a bit faster. The orders seem to be limit orders, fill or kill. I bought two lots of LTTs and one three-year T-bill; two went through immediately, the other timed out after about two minutes, then processed immediately once I put the order through again. It's much better than the old Fidelity bond trading platform, which was exasperating. And the spreads are narrow, so all in all an improvement.
Also, the Fidelity bond trading platform has been modernized and the problems Sophie mentioned are almost completely eliminated. You can now trade a single bond as easily as 10 or more, knowing what the price will be. The only drawback is that you have to be quick with placing the trade: refresh the page to get the most recent bids/asks, select the bond based on minimum trade (there's a list of maybe six different lots available for trading, down to a single bond), enter the number of bonds to be traded, and go! But, in my experience, it must all be done within about 10 seconds, otherwise an error message might appear saying that the prices have changed. It's not too difficult to go back and make the selections again, just be a bit faster. The orders seem to be limit orders, fill or kill. I bought two lots of LTTs and one three-year T-bill; two went through immediately, the other timed out after about two minutes, then processed immediately once I put the order through again. It's much better than the old Fidelity bond trading platform, which was exasperating. And the spreads are narrow, so all in all an improvement.
-
mathjak107
- Executive Member
- Posts: 4456
- Joined: Fri Jun 19, 2015 2:54 am
- Location: bayside queens ny
- Contact:
Re: tax-loss harvesting question
i gave up trying to deal with buying bonds like that at fidelity …it was a waste of time trying to beat the clock and after trade after trade was killed i said forget it
Re: tax-loss harvesting question
What would be the best replacement for SHY to buy and keep for 30 days to avoid wash sale at Fidelity?
Sold some SHY for tax-loss harvesting.
Probably just keeping the money in the settlement fund is good enough?
Sold some SHY for tax-loss harvesting.
Probably just keeping the money in the settlement fund is good enough?
