Using up large accumulated capital loss with treasury ETFs

[02nz]

OP,
Brilliant strategy on the upside of interest rates.
But as the rates are declining now it seems to be less effective moving forward, no?

Yes of course the amount of the benefit is directly proportional to interest income on the ETFs.

[sycamore]

I manage investments for a relative, who realized (before my taking over) huge capital losses in a single stock, mid-7 figures...
...
I've been thinking about how to "use up" these accumulated losses.

02nz, thanks for starting this thread about 13 months ago. It's an interesting solution to a difficult problem.

Can you share how much progress you've made at using up the losses? Is your relative happy with your approach?

[hiddenpower]

Awesome thanks! Are ex dates traditionally so reliable or does it depend on the underlying?

I'm not sure what you mean by "does it depend on the underlying." The published dates are 100% reliable.

In general are ex-dividend dates usually published so far in advance or do they sometimes get published with short notice? Seems like it's pre-published for tbill etfs far in advance but maybe not for other types of etfs and stocks?

[02nz]

Awesome thanks! Are ex dates traditionally so reliable or does it depend on the underlying?

I'm not sure what you mean by "does it depend on the underlying." The published dates are 100% reliable.

In general are ex-dividend dates usually published so far in advance or do they sometimes get published with short notice? Seems like it's pre-published for tbill etfs far in advance but maybe not for other types of etfs and stocks?

I'm puzzled why you're still asking this when I already linked the published schedule for the ex-dividend dates for SGOV through the end of 2025.

viewtopic.php?p=8045337#p8045337

[hiddenpower]

Awesome thanks! Are ex dates traditionally so reliable or does it depend on the underlying?

I'm not sure what you mean by "does it depend on the underlying." The published dates are 100% reliable.

In general are ex-dividend dates usually published so far in advance or do they sometimes get published with short notice? Seems like it's pre-published for tbill etfs far in advance but maybe not for other types of etfs and stocks?

I'm not sure why you're still asking this when I already linked the published schedule for the ex-dividend dates for SGOV through the end of 2025.

viewtopic.php?p=8045337#p8045337

Because my question isn't being understood. My question was adjacent on if ex-dividends are always regularly predictable and set in other cases, say for non-tbill-etfs. I was just curious.

[hiddenpower]

And for posterity, I found two solutions for automated alerts on ex-dividend announcement date:
thedividendtracker.com and divtracker app on iphone.

[whodidntante]

Buying a box spread would be easier IMO.

[02nz]

I manage investments for a relative, who realized (before my taking over) huge capital losses in a single stock, mid-7 figures...
...
I've been thinking about how to "use up" these accumulated losses.

02nz, thanks for starting this thread about 13 months ago. It's an interesting solution to a difficult problem.

Can you share how much progress you've made at using up the losses? Is your relative happy with your approach?

We haven't been focused on "progress" toward using up the entirety of the accumulated loss, he'll never use up the entire amount during his lifetime. But with about $2M in fixed income, 5%+ yield in SGOV (at least for now), and a nearly 50% marginal tax rate state + federal, that's $50K+ in tax savings annually.

[02nz]

Awesome thanks! Are ex dates traditionally so reliable or does it depend on the underlying?

I'm not sure what you mean by "does it depend on the underlying." The published dates are 100% reliable.

In general are ex-dividend dates usually published so far in advance or do they sometimes get published with short notice? Seems like it's pre-published for tbill etfs far in advance but maybe not for other types of etfs and stocks?

I'm not sure why you're still asking this when I already linked the published schedule for the ex-dividend dates for SGOV through the end of 2025.

viewtopic.php?p=8045337#p8045337

Because my question isn't being understood. My question was adjacent on if ex-dividends are always regularly predictable and set in other cases, say for non-tbill-etfs. I was just curious.

I suggest you start a separate thread then.

[hiddenpower]

I'm not sure what you mean by "does it depend on the underlying." The published dates are 100% reliable.

In general are ex-dividend dates usually published so far in advance or do they sometimes get published with short notice? Seems like it's pre-published for tbill etfs far in advance but maybe not for other types of etfs and stocks?

I'm not sure why you're still asking this when I already linked the published schedule for the ex-dividend dates for SGOV through the end of 2025.

viewtopic.php?p=8045337#p8045337

Because my question isn't being understood. My question was adjacent on if ex-dividends are always regularly predictable and set in other cases, say for non-tbill-etfs. I was just curious.

I suggest you start a separate thread then.

Rawr back at you.

[ejm009]

I messed up and missed Aug=>Sep sell and re-buy.
Did not account for Labor day market closure.
Here are my remaining target dates for the year.

Ex-Date (re-Buy) Ex-Day Sell Date Sell Day Note
1-Oct-24 Tue 30-Sep Mon
1-Nov-24 Fri 31-Oct Thu
2-Dec-24 Mon 29-Nov Fri Early
18-Dec-24 Wed 17-Dec Tue Extra
30-Dec-24 Mon 27-Dec Fri

Market closes early day after Thanksgiving.
If you see an error, please flag.

[invstar]

I have a related question. I am also planning to follow the same approach of focusing on capital gains instead of monthly dividends to offset my approximately $80k in capital losses. I bought SGOV on August 22nd but couldn’t sell it at the end of August, so I received dividends, which is fine. Now, if I sell it on September 30th, will I still see any profit? I understand I won't receive dividends since I’d be selling before the ex-dividend date, but my concern is that I bought the the shares when the price was relatively high in August (around $100.60). If the price on September 30th is only around $100.70, that’s just a $0.10 gain per share, while the dividend would have been about $0.40 per share. How would I generate any meaningful profit in this scenario if I sell on the 30th and then repurchase on October 1st? Thanks.

[dratkinson]

I have a related question. I am also planning to follow the same approach of focusing on capital gains instead of monthly dividends to offset my approximately $80k in capital losses. I bought SGOV on August 22nd but couldn’t sell it at the end of August, so I received dividends, which is fine. Now, if I sell it on September 30th, will I still see any profit? I understand I won't receive dividends since I’d be selling before the ex-dividend date, but my concern is that I bought the the shares when the price was relatively high in August (around $100.60). If the price on September 30th is only around $100.70, that’s just a $0.10 gain per share, while the dividend would have been about $0.40 per share. How would I generate any meaningful profit in this scenario if I sell on the 30th and then repurchase on October 1st? Thanks.

Forget your first purchase date and smaller CG. Why?  You must sell and get a profit at the end of the month.  Why?  So you can continue/start over.


#1. If you sell for a profit. You can continue/start over.  
--You must sell before next ex-dividend date; as long as it's a profit, wash-sale rules don't apply, and you're back to owning zero shares.
--Once you own zero shares, then you can continue/start over to: buy on/after* next record date, and sell before next ex-dividend date. (* As you now understand, the sooner the better.)


#2. If you don't sell. Delay selling for one month.
--If you forget to sell in a month (it's on/after current ex-dividend date), you'll get the distro and 1099DIV.  
--Don't sell until the trading day before the next ex-dividend date.  This will give you another profit sale and again start you over owning zero shares.


#3. If you sell for a loss. Don't sell for a loss, as that invokes wash-sale holding-period requirements and recordkeeping/tax reporting hassle.
--Your broker should handle the recordkeeping for you and send you correct tax documents.
--But it's your responsibility to ensure those documents are reported correctly on your tax return, which means you must learn/do the same cost-basis tracking to correct the broker's tax documents if wrong.  (Remember, you're only buying/selling one tax lot, every month, and dates are reported on 1099B to IRS; if IRS computers see a wash sale and you didn't report it correctly ...you'll read about it.)
--It's much simpler to avoid the hassle than it is to deal with it.  So to avoid it, only sell when you have a profit, and your largest profit should occur on the trading day before the next ex-dividend date.

Worst case. If you sell for a loss, then you have a TLH and wash-sale holding periods apply.
--Don't repurchase for >30days to allow wash-sale window to close.
--If >30day delay puts you near next ex-dividend date*, then consider delaying next purchase until on/after next record date. (* Like your current situation, same work to buy/sell but smaller CG if you buy late in the month. Your choice.)
See: https://www.bogleheads.org/wiki/Tax_loss_harvesting



Edit.  I think that covers the high point of everything. Clarified sale transaction cases: profit, didn't sell, loss.

[invstar]

Thanks dratkinson. Very useful info.

So, I should sell all my shares on 30th September and then buy again on 1st October to ensure I get profit and no taxable dividends. Is this assumption right? Based on the linked posted above, the EX-DATE and RECORD DATE both fall on 1st October.

Also, since I am not worried about TLH and just want to realize gains and offset them with my existing capital loss - does-wash sale matter?

Thanks.

I have a related question. I am also planning to follow the same approach of focusing on capital gains instead of monthly dividends to offset my approximately $80k in capital losses. I bought SGOV on August 22nd but couldn’t sell it at the end of August, so I received dividends, which is fine. Now, if I sell it on September 30th, will I still see any profit? I understand I won't receive dividends since I’d be selling before the ex-dividend date, but my concern is that I bought the the shares when the price was relatively high in August (around $100.60). If the price on September 30th is only around $100.70, that’s just a $0.10 gain per share, while the dividend would have been about $0.40 per share. How would I generate any meaningful profit in this scenario if I sell on the 30th and then repurchase on October 1st? Thanks.

Forget your first purchase date and smaller CG. Why?  You must sell and get a profit at the end of the month.  Why?  

#1. So you can start over.  
--You must sell before (first of) Oct's ex-dividend date; as long as it's a profit, wash-sale rules don't apply, and you're back to owning zero shares.
--Once you own zero shares, then you've started over to: buy on/after next record date (as you've learned, the sooner the better), and sell before next ex-dividend date.


#2. If you forget to sell in a month (ex-dividend date has passed), you'll get the distro and 1099DIV.  But don't sell until the trading day before the next ex-dividend date.  This will give you another profit sale and again start you over owning zero shares.


#3. You don't want to sell for a loss, as that will invoke the wash-sale holding-period requirements and recordkeeping/tax reporting hassle.
--Your broker should handle the recordkeeping for you and send you correct tax documents.
--But it's your responsibility to ensure those documents are reported correctly on your tax return, which means you must do the same recordkeeping to correct the broker's documents if wrong.  (Remember, you're only buying/selling one tax lot, every month, and dates are reported on 1099B to IRS; if its computers see a wash sale and you didn't report it correctly ...you'll read about it.)
--It's much simpler to avoid the hassle than it is to fix it.  So to avoid it, only sell when you have a profit, and your largest profit should occur on the trading day before the ex-dividend date.

Worst case. If you sell for a loss, then you have a TLH and wash-sale holding periods apply.
--Don't rebuy for >30days to allow wash-sale window to close.
--If you are then near the next ex-dividend date, then delay buying until on/after the next record date.
See: https://www.bogleheads.org/wiki/Tax_loss_harvesting



Edit.  I think that covers everything.

[Rocky Mtn Man]

Also, since I am not worried about TLH and just want to realize gains and offset them with my existing capital loss - does-wash sale matter?

Thanks.

No. Wash-sales only apply to losses. You will always have a short term capital gain when following this strategy. It may help to look at a graph of the price of SGOV or BIL. It looks like a sawtooth. You buy in the valley and sell at the peak, so it is always a capital gain.

[dratkinson]

...
Also, since I am not worried about TLH and just want to realize gains and offset them with my existing capital loss - does-wash sale matter?

My bad, I should have been more clear. I was trying to be complete, so tried to hit the high points of what I believed were the expected transaction outcomes: you sold for a profit, you didn't sell, you sold for a loss.

So, if you sell for a profit, then the other outcomes/cases don't matter to you. But if some day they do matter to you, then they are covered.

[invstar]

Thanks dratkinson.

...
Also, since I am not worried about TLH and just want to realize gains and offset them with my existing capital loss - does-wash sale matter?

My bad, I should have been more clear. I was trying to be complete, so tried to hit the high points of what I believed were the expected transaction outcomes: you sold for a profit, you didn't sell, you sold for a loss.

So, if you sell for a profit, then the other outcomes/cases don't matter to you. But if some day they do matter to you, then they are covered.

[invstar]

Thanks Rocky Mtn Man.

Also, since I am not worried about TLH and just want to realize gains and offset them with my existing capital loss - does-wash sale matter?

Thanks.

No. Wash-sales only apply to losses. You will always have a short term capital gain when following this strategy. It may help to look at a graph of the price of SGOV or BIL. It looks like a sawtooth. You buy in the valley and sell at the peak, so it is always a capital gain.

[02nz]

So, I should sell all my shares on 30th September and then buy again on 1st October to ensure I get profit and no taxable dividends. Is this assumption right? Based on the linked posted above, the EX-DATE and RECORD DATE both fall on 1st October.

Yes, sell before market close on Sep 30 and buy after market close open on Oct 1.

[invstar]

So, I should sell all my shares on 30th September and then buy again on 1st October to ensure I get profit and no taxable dividends. Is this assumption right? Based on the linked posted above, the EX-DATE and RECORD DATE both fall on 1st October.

Yes, sell before market close on Sep 30 and buy after market close on Oct 1.

Thanks 02nz. I purchased it for the first time on August 22nd and price was 100.60. On September 3rd, the price dropped to 100.34 and dividends were paid. If I sell my shares on September 30th for 100.73, what will be the cost basis for my shares? Will it be 100.60 (original purchased price) OR 100.34 (beginning of the month's reset price)?
I want to make sure that I won't loose anything compared to holding until the ex-dividend date to get the dividends. Thanks.

and Why not buy on Oct 1st during market hours and why after market hours?

[bongo]

So, I should sell all my shares on 30th September and then buy again on 1st October to ensure I get profit and no taxable dividends. Is this assumption right? Based on the linked posted above, the EX-DATE and RECORD DATE both fall on 1st October.

Yes, sell before market close on Sep 30 and buy after market close on Oct 1.

Just eyeballing https://finance.yahoo.com/quote/SGOV/history/?ltr=1 seems like this is a bit lossy. If you got the closing price the day before and re-bought at the open, it's about a 8-10% loss compared to the div. Of course, the tax savings more than makes up for it, but I wonder if there's a way to mitigate the loss. On 100k gain, it's about -8k a year.

Code: Select all

		Open	Close
Sep 3, 2024	0.44 Dividend
Sep 3, 2024	100.34	100.34
Aug 30, 2024	100.74	100.74	(= Cap gain .40)

Aug 1, 2024	0.45 Dividend
Aug 1, 2024	100.30	100.30
Jul 31, 2024	100.72	100.72	(= Cap gain .42)

I guess being out of the fund 1/30th of the days accounts for about 3%. Then a 1cent bid/ask may be another 2.5%. Not sure where the rest of the loss goes, maybe if out over weekends? I wonder if using a trading partner with a different ex-div date would help. Eg USFR is about a week before, and then you also have the freedom to swap same-day anytime, as long as you avoid both divs.

[sycamore]

Thanks 02nz. I purchased it for the first time on August 22nd and price was 100.60. On September 3rd, the price dropped to 100.34 and dividends were paid. If I sell my shares on September 30th for 100.73, what will be the cost basis for my shares? Will it be 100.60 (original purchased price) OR 100.34 (beginning of the month's reset price)?

Here is the relevant part: "I purchased it for the first time on August 22nd and price was 100.60"

So your cost basis for those shares is 100.60.

Every separate purchase has its own cost basis and holding period. Doesn't matter if dividends were paid on the shares.

Did you reinvest the dividends? If so, that counts as a separate purchase, and would have its own cost basis (the price at reinvestment) and holding period.

[02nz]

Why not buy on Oct 1st during market hours and why after market hours?

Sorry, that was a typo, you would buy on October 1 any time after the market open. Fixed above.

[02nz]

So, I should sell all my shares on 30th September and then buy again on 1st October to ensure I get profit and no taxable dividends. Is this assumption right? Based on the linked posted above, the EX-DATE and RECORD DATE both fall on 1st October.

Yes, sell before market close on Sep 30 and buy after market close on Oct 1.

Just eyeballing https://finance.yahoo.com/quote/SGOV/history/?ltr=1 seems like this is a bit lossy. If you got the closing price the day before and re-bought at the open, it's about a 8-10% loss compared to the div. Of course, the tax savings more than makes up for it, but I wonder if there's a way to mitigate the loss. On 100k gain, it's about -8k a year.

Code: Select all

		Open	Close
Sep 3, 2024	0.44 Dividend
Sep 3, 2024	100.34	100.34
Aug 30, 2024	100.74	100.74	(= Cap gain .40)

Aug 1, 2024	0.45 Dividend
Aug 1, 2024	100.30	100.30
Jul 31, 2024	100.72	100.72	(= Cap gain .42)

I guess being out of the fund 1/30th of the days accounts for about 3%. Then a 1cent bid/ask may be another 2.5%. Not sure where the rest of the loss goes, maybe if out over weekends? I wonder if using a trading partner with a different ex-div date would help. Eg USFR is about a week before, and then you also have the freedom to swap same-day anytime, as long as you avoid both divs.

Originally I considered using a "trade partner" but decided against it because 1) it would mean twice the work; 2) there are relatively few short-term treasury ETFs that don't distribute on the first of the month; 3) those few that fit the bill often have larger intra-day variations in the NAV, probably because they don't have SGOV's volume; 4) for the one night (or 3 nights over a weekend) a month we are out of the market, we get interest on the cash from the brokerage at comparable rates (but without the tax savings of course); and 5) an extra buy/sell cycle every month would mean paying the SEC fee (0.00278% of the sale) twice.

[dratkinson]

^^^ The simplicity of using only one fund is appealing.



Trade partners. Was wondering what might work better if SGOV fell out of favor after interest rates revert to normal.

Searched "ultra short-term treasury ETF": https://www.google.com/search?&q=ultra+ ... easury+etf

Found this list: https://www.etf.com/topics/ultra-short-term
BIL is on list before SGOV.


Example from above list: BIL; see: https://www.ssga.com/us/en/intermediary ... ll-etf-bil
--3mo NAV chart (click) shows embedded dividends, indicates ~EOM ex-date: https://www.morningstar.com/etfs/arcx/bil/chart
--Daily bid/asked spread seems to be 1-cent: https://www.morningstar.com/etfs/ARCX/BIL/quote (Looking for low volatility.)
--Avg daily trading volume: ~7M (Looking for enough volume to indicate a liquid investment. Question: what should we consider illiquid?)
--ER: 0.1356% (Cost is one of the few things we can control.)
--SEC yield: 5.04% (More is better, but maybe not at the expense of higher bid/asked spread and liquidity; not the case with BIL, in general.)


Question. If we want to search for trade partners, does above cover the parameters necessary to decide for/against a trade partner?

Before execution, once we winnow a list of trade partners, we'd want to find/read prospectus/distribution schedule before making a final selection and building a tax-gain harvest game plan.

Question. Did I miss anything?

[trueblueky]

Any thoughts on using margin to leverage this?

[02nz]

Was wondering what might work better if SGOV fell out of favor after interest rates revert to normal.

Searched "ultra short-term treasury ETF": https://www.google.com/search?&q=ultra+ ... easury+etf

Found this list: https://www.etf.com/topics/ultra-short-term
BIL is on list before SGOV.

What do you think makes BIL potentially better than SGOV? BIL yields a little less than SGOV, and when "interest rates revert to normal," both yields will fall, they have about the same duration.

[02nz]

Any thoughts on using margin to leverage this?

Maybe I'm misunderstanding you, but are you proposing to pay 6% interest on margin to get 5% yield on SGOV?

[dratkinson]

Was wondering what might work better if SGOV fell out of favor after interest rates revert to normal.

Searched "ultra short-term treasury ETF": https://www.google.com/search?&q=ultra+ ... easury+etf

Found this list: https://www.etf.com/topics/ultra-short-term
BIL is on list before SGOV.

What do you think makes BIL potentially better than SGOV? BIL yields a little less than SGOV, and when "interest rates revert to normal," both yields will fall, they have about the same duration.

I don't believe BIL to be better than SGOV ...just used it as an example of my thinking.

From the linked list, there were some non-treasury options; was thinking that when/if treasuries drop, then other options might drop less; and if so, what parameters should interest us to winnow them, then.

[trueblueky]

Any thoughts on using margin to leverage this?

Maybe I'm misunderstanding you, but are you proposing to pay 6% interest on margin to get 5% yield on SGOV?

Is that the current rate? I don't do margin. Would it ever make sense to use it in this particular circumstance?

[dratkinson]

Any thoughts on using margin to leverage this?

Maybe I'm misunderstanding you, but are you proposing to pay 6% interest on margin to get 5% yield on SGOV?

Is that the current rate? I don't do margin. Would it ever make sense to use it in this particular circumstance?

If, after using margin, your after-tax income were expected to drop below what your invested principal could produce otherwise, it would be a waste of principal to use margin. You'd need to research your numbers before proceeding; should be easy to do with your broker.

A quick check at mine says Vanguard's margin base rate is ~10.75%, and can change at any time: https://investor.vanguard.com/client-benefits/margin
No thanks.

Plus it's another complication ...added to trying to remember to buy/sell on time.



A positive thought. Since 02nz's method is (currently) producing ~5%, exempt from both fed+state tax, I believe it's fair to compute a taxable-equivalent yield for it the same way we do a single-state muni fund.

Example: in my 22%/5% fed/state tax brackets, its TEY = 6.85% (5%/(1 -.22 -.05)). So before I do monthly rebalancing, I believe that's the number I should compare against my other fixed-income alternatives.

[EddyB]

I messed up and missed Aug=>Sep sell and re-buy.
Did not account for Labor day market closure.
Here are my remaining target dates for the year.

Ex-Date (re-Buy) Ex-Day Sell Date Sell Day Note
1-Oct-24 Tue 30-Sep Mon
1-Nov-24 Fri 31-Oct Thu
2-Dec-24 Mon 29-Nov Fri Early
18-Dec-24 Wed 17-Dec Tue Extra
30-Dec-24 Mon 27-Dec Fri

Market closes early day after Thanksgiving.
If you see an error, please flag.

I have done this a couple of times at Schwab, but did it on IBKR today, where I noticed that I can post an order now to only be submitted at a scheduled future time. I set a market sale to be submitted at9:35 a.m. on the last day of this month. I have a lot of time to cancel it, but has anyone else tried this or have any special concerns?

[02nz]

I messed up and missed Aug=>Sep sell and re-buy.
Did not account for Labor day market closure.

How would you have missed the Aug/Sep "cycle" because of Labor Day, which was Monday, September 2? You would have sold on Friday, Aug 30 regardless; the market was closed on September 2, so you would've just bought on Tuesday, September 3.

[lgerla]

Thank you 02nz for posting this strategy. I have a question about Dec. To capture December's gains, I would :

Sell on Nov 30 (or Dec 1), Buy on ex-div date of Dec 2.
Sell on Dec. 17, buy on ex-div date of Dec. 18.
Then hold, sell on Jan. 31, Feb. 1, or Feb. 2, 2025 and buy on Feb. 3

Is that correct?

I see on the schedule the notation "Potential Income Distribution for Excise Purposes" on Dec. 30, so wondered if I would sell on Dec. 29 to see if NAV drops on Dec. 30.

Thanks again.

[02nz]

Thank you 02nz for posting this strategy. I have a question about Dec. To capture December's gains, I would :

Sell on Nov 30 (or Dec 1), Buy on ex-div date of Dec 2.
Sell on Dec. 17, buy on ex-div date of Dec. 18.
Then hold, sell on Jan. 31, Feb. 1, or Feb. 2, 2025 and buy on Feb. 3

Is that correct?

I see on the schedule the notation "Potential Income Distribution for Excise Purposes" on Dec. 30, so wondered if I would sell on Dec. 29 to see if NAV drops on Dec. 30.

Thanks again.

Every month except for December, I sell on the last trading day of the calendar month, and rebuy on the first trading day of the next month. I'm not 100% sure what happens if you sell outside market hours before the ex-dividend date, e.g., on Sunday, Dec 1, but I would avoid that anyway because bid/ask spreads will be wider outside market hours.

That means I will sell on Nov 29 (Fri - and markets will close early because it's the day after Thanksgiving), rebuy on Dec 2 (Mon). Similarly, I will sell Jan 31 (Fri) and rebuy Feb 3 (Mon). For the "extra" mid-December distribution, I will sell Dec 17 and rebuy on the ex-dividend date of Dec 18.

The "potential distribution" on Dec 30 has never actually occurred in the history of the fund.

[ejm009]

I missed the Aug=>Sept as I knew the ex-date was Tuesday the 3rd and thought I could then sell on Monday the 2nd. I did not sell on Friday and market was closed by the time I realized my error.

[02nz]

I missed the Aug=>Sept as I knew the ex-date was Tuesday the 3rd and thought I could then sell on Monday the 2nd. I did not sell on Friday and market was closed by the time I realized my error.

You and lgerla both got confused by the ex-dividend dates, which aren't always on the 1st of the month because of weekends/holidays. A simpler way to think about it is: except in December (which has a distribution on the middle of the month), you ALWAYS need to sell on the last trading day of the calendar month, and rebuy on the first trading day of the following month, which is the ex-dividend date.

I have a reminder on my phone. Once I've done the selling, I change it to the last trading day of the following month.

[ejm009]

I like your last trading day of the month rule !
Thank you.