Let's Talk SPX Box Spreads

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Let's Talk SPX Box Spreads

Post by comeinvest »

indexfundfan wrote: Tue May 09, 2023 3:36 pm I'm pretty sure treasuries settle on day T on maturity.
That would make the T-bill yield even higher, and the discrepancy per my post between T-bill yield and Term SOFR as well as between T-bill and my SPX spread even bigger. Term SOFR is usually higher than corresponding T-bill yields. T+2 settlement of T-bills at maturity would make the yield numbers more consistent. The purchase of the T-bill settles on T+1 in retail broker accounts, according to other threads in this forum. I couldn't find easy information about the settlement at maturity.
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Let's Talk SPX Box Spreads

Post by comeinvest »

Perhaps related to my question in my previous comments: Why are both BID and ASK yields of all the May 2023 T-bill expirations below 5%, all the while the fed funds rate and all other reference rates are currently above 5% ?

Then yields jump by about 1.4 percentage points in the 2 days between May 30 and June 01.

Image
User avatar
Kevin M
Posts: 15789
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Let's Talk SPX Box Spreads

Post by Kevin M »

You should have access to proceeds from a maturing Treasury on the maturity date if it is a trading day. However, different brokers handle it differently with respect to showing the cash as settled. Vanguard is slower than Fidelity.

However, we have found that you can usually buy Treasuries with settlement on the maturity date with the proceeds from the maturing Treasury even if it doesn't show the proceeds as settled cash. So, for maturity on a Friday, you should be able to place a secondary Treasury purchase order on the day before (Thursday). You also should be able to place an auction order with issue date of new Treasury = maturity date of maturing Treasury.
If I make a calculation error, #Cruncher probably will let me know.
User avatar
Kevin M
Posts: 15789
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Let's Talk SPX Box Spreads

Post by Kevin M »

comeinvest wrote: Tue May 09, 2023 4:21 pm Perhaps related to my question in my previous comments: Why are both BID and ASK yields of all the May 2023 T-bill expirations below 5%, all the while the fed funds rate and all other reference rates are currently above 5% ?

Then yields jump by about 1.4 percentage points in the 2 days between May 30 and June 01.
Yes.

Image

Some of us are pretty sure this is related to the June 1 debt ceiling deadline.
If I make a calculation error, #Cruncher probably will let me know.
User avatar
Kevin M
Posts: 15789
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Let's Talk SPX Box Spreads

Post by Kevin M »

The chart is for ask yields.
If I make a calculation error, #Cruncher probably will let me know.
parval
Posts: 153
Joined: Tue Oct 22, 2019 9:23 pm

Re: Let's Talk SPX Box Spreads

Post by parval »

In a taxable, how does BIL compare with a 1 to 3 month box spread?

I think the yield is almost the same, but BIL pays short term whereas the box is 60/40 right?
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Let's Talk SPX Box Spreads

Post by comeinvest »

Kevin M wrote: Tue May 09, 2023 7:08 pm
comeinvest wrote: Tue May 09, 2023 4:21 pm Perhaps related to my question in my previous comments: Why are both BID and ASK yields of all the May 2023 T-bill expirations below 5%, all the while the fed funds rate and all other reference rates are currently above 5% ?

Then yields jump by about 1.4 percentage points in the 2 days between May 30 and June 01.
...
Some of us are pretty sure this is related to the June 1 debt ceiling deadline.
Sorry if I missed the discussion, but what is the logic? The T-bills with maturities before June have yields significantly below the current fed funds rate (5 to 5.25%). What does this have to do with the debt ceiling? The anomaly seems to be the May maturities having yields too low, not the June maturities having yields too high.
Yes it might explain the June expiration having slightly higher yields than the fed funds rate; but it doesn't explain the May expiration having yields significantly below fed funds rate.

Yes it might explain the June maturities having higher yields than Term SOFR and LIBOR, and yields about equal to those of options spreads.
Last edited by comeinvest on Tue May 09, 2023 10:43 pm, edited 3 times in total.
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Let's Talk SPX Box Spreads

Post by comeinvest »

Kevin M wrote: Tue May 09, 2023 7:01 pm You should have access to proceeds from a maturing Treasury on the maturity date if it is a trading day. However, different brokers handle it differently with respect to showing the cash as settled. Vanguard is slower than Fidelity.

However, we have found that you can usually buy Treasuries with settlement on the maturity date with the proceeds from the maturing Treasury even if it doesn't show the proceeds as settled cash. So, for maturity on a Friday, you should be able to place a secondary Treasury purchase order on the day before (Thursday). You also should be able to place an auction order with issue date of new Treasury = maturity date of maturing Treasury.
If you are saying that purchase has T+1 settlement and maturity has T+0 settlement, then you would have to subtract 1 day for the time length in the economic APY calc. I don't think IB does that in TWS.

You have to define what you mean by "show the proceeds as settled cash" on T+0. I assume you mean it does actually settle on T+0, not just show as settled (or show without indication of settlement) on T+0. An important distinction, because many brokers and banks don't show or differentiate between transaction dates and value (settlement) dates of transactions in the customer facing applications. Fidelity had some issues when I was a customer; Schwab and IB show the settled and unsettled account balances separately.

I don't know about notes and bonds, but bills only mature on Tuesdays and Thursdays, not Fridays. I would like to know if I would have to buy the next bill on T-1 or on T, the maturity date, in order to match the value dates (cash settlement dates for purpose of interest rate calculation on broker cash). Also for purpose of calculating the economic APY of T-bills from the day that I purchase it. I would appreciate if somebody can relate first-hand, verified experience, or a reference. Thanks in advance.
User avatar
Kevin M
Posts: 15789
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Let's Talk SPX Box Spreads

Post by Kevin M »

comeinvest wrote: Tue May 09, 2023 9:28 pm
Kevin M wrote: Tue May 09, 2023 7:08 pm
comeinvest wrote: Tue May 09, 2023 4:21 pm Perhaps related to my question in my previous comments: Why are both BID and ASK yields of all the May 2023 T-bill expirations below 5%, all the while the fed funds rate and all other reference rates are currently above 5% ?

Then yields jump by about 1.4 percentage points in the 2 days between May 30 and June 01.
...
Some of us are pretty sure this is related to the June 1 debt ceiling deadline.
Sorry if I missed the discussion, but what is the logic? The T-bills with maturities before June have yields significantly below the current fed funds rate (5 to 5.25%). What does this have to do with the debt ceiling? The anomaly seems to be the May maturities having yields too low, not the June maturities having yields too high.
Yes it might explain the June expiration having slightly higher yields than the fed funds rate; but it doesn't explain the May expiration having yields significantly below fed funds rate.

Yes it might explain the June maturities having higher yields than Term SOFR and LIBOR, and yields about equal to those of options spreads.
So, you think that it's a coincidence that the big spike between May 31 and June 1 maturities happens to be the day before and the day when Treasury supposedly can't pay all of its bills? Ha!
If I make a calculation error, #Cruncher probably will let me know.
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Let's Talk SPX Box Spreads

Post by comeinvest »

Kevin M wrote: Wed May 10, 2023 3:40 am
comeinvest wrote: Tue May 09, 2023 9:28 pm
Kevin M wrote: Tue May 09, 2023 7:08 pm
comeinvest wrote: Tue May 09, 2023 4:21 pm Perhaps related to my question in my previous comments: Why are both BID and ASK yields of all the May 2023 T-bill expirations below 5%, all the while the fed funds rate and all other reference rates are currently above 5% ?

Then yields jump by about 1.4 percentage points in the 2 days between May 30 and June 01.
...
Some of us are pretty sure this is related to the June 1 debt ceiling deadline.
Sorry if I missed the discussion, but what is the logic? The T-bills with maturities before June have yields significantly below the current fed funds rate (5 to 5.25%). What does this have to do with the debt ceiling? The anomaly seems to be the May maturities having yields too low, not the June maturities having yields too high.
Yes it might explain the June expiration having slightly higher yields than the fed funds rate; but it doesn't explain the May expiration having yields significantly below fed funds rate.

Yes it might explain the June maturities having higher yields than Term SOFR and LIBOR, and yields about equal to those of options spreads.
So, you think that it's a coincidence that the big spike between May 31 and June 1 maturities happens to be the day before and the day when Treasury supposedly can't pay all of its bills? Ha!
I didn't say that. I'm trying to find an explanation for the May expirations having APYs significantly below fed fund rates and other reference rates. I got 5.4% with SPX options; so T-bills at ca. 5.0-5.1% would be the normal spread, which would also be within the fed funds target rate. I can't explain the 4.1% rates for May expirations when current fed fund target range is 5-5.25%.

B.t.w. mid May options currently trade at ca. 5.7% APY - an unusual big spread to treasuries, even to the mid of the fed funds target range.

Image
MrJedi
Posts: 3540
Joined: Wed May 06, 2020 11:42 am

Re: Let's Talk SPX Box Spreads

Post by MrJedi »

I assume there is a premium on the treasuries that will mature before a potential default. Those investors want their money on time and don't want to risk a late payment.
User avatar
Kevin M
Posts: 15789
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Let's Talk SPX Box Spreads

Post by Kevin M »

MrJedi wrote: Wed May 10, 2023 9:24 am I assume there is a premium on the treasuries that will mature before a potential default. Those investors want their money on time and don't want to risk a late payment.
Right. Higher demand for issues maturing before June 1 -> lower yield for those issues.
If I make a calculation error, #Cruncher probably will let me know.
tj-longterm
Posts: 592
Joined: Tue Apr 10, 2007 10:10 am

Re: Let's Talk SPX Box Spreads

Post by tj-longterm »

I bought sold a small amount of Dec '27 box spreads as an experiment to see how it would work at TD Ameritrade. It was easy to do, but I've been surprised that outside of trading hours, TD Ameritrade very frequently marks my box spreads up to a gain of 30%, which then disappears once trading opens again. It's very frequent, and it's in my favor (which is good for margin call risk), but makes me wonder about year end tax reporting.

Anyone have any idea how the value will be reported by a broker like TD Ameritrade at year end?

In the end, it shouldn't matter too much, especially since 1256 losses can be carried back 3 years against prior 1256 gains, but I'm wondering what will be considered "fair market" value at year end.
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Let's Talk SPX Box Spreads

Post by comeinvest »

tj-longterm wrote: Mon May 15, 2023 8:44 am I bought sold a small amount of Dec '27 box spreads as an experiment to see how it would work at TD Ameritrade. It was easy to do, but I've been surprised that outside of trading hours, TD Ameritrade very frequently marks my box spreads up to a gain of 30%, which then disappears once trading opens again. It's very frequent, and it's in my favor (which is good for margin call risk), but makes me wonder about year end tax reporting.

Anyone have any idea how the value will be reported by a broker like TD Ameritrade at year end?

In the end, it shouldn't matter too much, especially since 1256 losses can be carried back 3 years against prior 1256 gains, but I'm wondering what will be considered "fair market" value at year end.
Off hours is probably meaningless, especially for tax reporting.
What implied rate did you get?
petulant
Posts: 3601
Joined: Thu Sep 22, 2016 1:09 pm

Re: Let's Talk SPX Box Spreads

Post by petulant »

I think TJ's worry is whether the broker will use the market quotes from before close on the last business day of the year, or whether it will use the outsized off-hours quotes, when it calculates the mark-to-market values.
Kbg
Posts: 355
Joined: Thu Mar 23, 2017 11:33 am

Re: Let's Talk SPX Box Spreads

Post by Kbg »

I can't imagine the tax form mark to market value wouldn't be based on the official close of the last trading day of the year for the respective instrument. I'm not a tax lawyer, but I think it would be hard to justify else wise to the IRS.
petulant
Posts: 3601
Joined: Thu Sep 22, 2016 1:09 pm

Re: Let's Talk SPX Box Spreads

Post by petulant »

Kbg wrote: Mon May 15, 2023 8:38 pm I can't imagine the tax form mark to market value wouldn't be based on the official close of the last trading day of the year for the respective instrument. I'm not a tax lawyer, but I think it would be hard to justify else wise to the IRS.
If I was TJ, which I am not, my worry would be more along the lines of what the broker reports on the 1099, not what can be justified to the IRS. The broker may or may not be taking a good position on behalf of its clients. Believe it or not, companies do dumb things (occasionally).
Kbg
Posts: 355
Joined: Thu Mar 23, 2017 11:33 am

Re: Let's Talk SPX Box Spreads

Post by Kbg »

petulant wrote: Mon May 15, 2023 9:48 pm
Kbg wrote: Mon May 15, 2023 8:38 pm I can't imagine the tax form mark to market value wouldn't be based on the official close of the last trading day of the year for the respective instrument. I'm not a tax lawyer, but I think it would be hard to justify else wise to the IRS.
If I was TJ, which I am not, my worry would be more along the lines of what the broker reports on the 1099, not what can be justified to the IRS. The broker may or may not be taking a good position on behalf of its clients. Believe it or not, companies do dumb things (occasionally).
And if not the official closing on the last day of the year, what else would a broker use?

The broker DOES have to justify what they use for tax reporting when they report basis...and it has to be the same across thousands of clients. They aren't going to pull a number out of thin air.
petulant
Posts: 3601
Joined: Thu Sep 22, 2016 1:09 pm

Re: Let's Talk SPX Box Spreads

Post by petulant »

Kbg wrote: Tue May 16, 2023 6:45 pm
petulant wrote: Mon May 15, 2023 9:48 pm
Kbg wrote: Mon May 15, 2023 8:38 pm I can't imagine the tax form mark to market value wouldn't be based on the official close of the last trading day of the year for the respective instrument. I'm not a tax lawyer, but I think it would be hard to justify else wise to the IRS.
If I was TJ, which I am not, my worry would be more along the lines of what the broker reports on the 1099, not what can be justified to the IRS. The broker may or may not be taking a good position on behalf of its clients. Believe it or not, companies do dumb things (occasionally).
And if not the official closing on the last day of the year, what else would a broker use?

The broker DOES have to justify what they use for tax reporting when they report basis...and it has to be the same across thousands of clients. They aren't going to pull a number out of thin air.
...TJ literally just posted about how the broker would report different numbers during off hours, which are presumably based on some kind of different model or bid-ask book. So the question is whether the broker would ever use this for the tax reporting. Wherever the broker gets these numbers, they are something, so his question is simple: can anybody verify that the broker uses numbers from close, not these off-hours numbers? Trying to resort to "tax law requires it" is not a solution to this question because, again, companies do stupid things. The question is can somebody operationally confirm the broker uses the market close numbers, not a Panglossian response based on confused notions of tax compliance.
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Let's Talk SPX Box Spreads

Post by comeinvest »

Nobody in this forum can guarantee you that the broker won't make some mistake at some time.
Call the broker (even then unlikely you get a guarantee), and/or wait until you get the form, complain if it's incorrect, and be done with it. It's a non-issue unless you actually receive an incorrect form. Before that happens, enjoy life.
ggtt
Posts: 23
Joined: Thu Jan 19, 2023 11:32 am

Re: Let's Talk SPX Box Spreads

Post by ggtt »

Kbg wrote: Tue May 16, 2023 6:45 pm
petulant wrote: Mon May 15, 2023 9:48 pm
Kbg wrote: Mon May 15, 2023 8:38 pm I can't imagine the tax form mark to market value wouldn't be based on the official close of the last trading day of the year for the respective instrument. I'm not a tax lawyer, but I think it would be hard to justify else wise to the IRS.
If I was TJ, which I am not, my worry would be more along the lines of what the broker reports on the 1099, not what can be justified to the IRS. The broker may or may not be taking a good position on behalf of its clients. Believe it or not, companies do dumb things (occasionally).
And if not the official closing on the last day of the year, what else would a broker use?

The broker DOES have to justify what they use for tax reporting when they report basis...and it has to be the same across thousands of clients. They aren't going to pull a number out of thin air.
It's not obvious how to mark for instruments which basically never trade (and I'm guessing most SPX Dec 27 options fall into this bucket). For example what if the option last traded 3 months ago and SPX moved 15% since then?

I don't know how brokers handle this, but I could imagine they do something different like use the midpoint on of the last bid/ask or delta adjust from the last trade price.
ggtt
Posts: 23
Joined: Thu Jan 19, 2023 11:32 am

Re: Let's Talk SPX Box Spreads

Post by ggtt »

Kbg wrote: Tue May 16, 2023 6:45 pm
petulant wrote: Mon May 15, 2023 9:48 pm
Kbg wrote: Mon May 15, 2023 8:38 pm I can't imagine the tax form mark to market value wouldn't be based on the official close of the last trading day of the year for the respective instrument. I'm not a tax lawyer, but I think it would be hard to justify else wise to the IRS.
If I was TJ, which I am not, my worry would be more along the lines of what the broker reports on the 1099, not what can be justified to the IRS. The broker may or may not be taking a good position on behalf of its clients. Believe it or not, companies do dumb things (occasionally).
And if not the official closing on the last day of the year, what else would a broker use?

The broker DOES have to justify what they use for tax reporting when they report basis...and it has to be the same across thousands of clients. They aren't going to pull a number out of thin air.
Another complication is that for spreads, individual legs of the spread can get reported at prices very far from a reasonable market value. For example say you buy synthetic stock (long call short put) as part of your box spread. There will be a very liquid market for the spread, because it's just based off of SPX which is very liquid, plus easy to determine adjustments for interest rates and dividends. Say you compute the fair value to be $9.99 and manage to buy the spread for $10.00.

First we need to think about how each individual leg of the spread gets reported. If the bid/ask spreads of the underlying call and put are small then there is not much room for error because they must get reported within the bid/ask spread, but for an expiration 4 years from now they might be wide. So maybe the Call has a bid/ask of $14/16 and the Put has a bid/ask of $4/6. How do you report each leg of the spread? In my experience it can be arbitrary - maybe they will report that you bought the call for $15.50 and sold the put for $5.50 - but lets just say for simplicity the exchange uses the midpoint of the bid/ask, so you buy the call for $15 and sell the put for $5.

You do this in the afternoon on 12/31 so you think your broker will show no gain or loss for tax purposes. But then at the very end of the day someone comes in places a market order for the Call, buying it for $16. If your broker uses the last trade price, it will show that your spread is worth $11. That's clearly unfair since your position is a synthetic long and SPX didn't moved 10% in the last afternoon of the year, but it's what you get from just using the last trading price.

Again, no idea how the brokers actually handle this, but there's not a perfect way to do it and there's definitely a risk of weird things happening.
kxl19
Posts: 248
Joined: Wed Mar 13, 2013 12:41 am

Re: Let's Talk SPX Box Spreads

Post by kxl19 »

I'm curious if there was a hypothetical arbitrage opportunity a few weeks ago when the box spread yield was lower than a similar maturity treasury? (assumes no US default)

Example: Box spread June maturity was about 5.3%
Mid-June Treasury: yielding 5.5%+

Seems like the arbitrage opportunity gets bigger if you consider "after-tax" cost of box spreads (assuming the losses can offset existing gains) and similar after-tax yield of a treasury (interest tax exempt)

Anything I'm missing from this logic?
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Let's Talk SPX Box Spreads

Post by comeinvest »

ggtt wrote: Wed May 17, 2023 10:08 am
Kbg wrote: Tue May 16, 2023 6:45 pm
petulant wrote: Mon May 15, 2023 9:48 pm
Kbg wrote: Mon May 15, 2023 8:38 pm I can't imagine the tax form mark to market value wouldn't be based on the official close of the last trading day of the year for the respective instrument. I'm not a tax lawyer, but I think it would be hard to justify else wise to the IRS.
If I was TJ, which I am not, my worry would be more along the lines of what the broker reports on the 1099, not what can be justified to the IRS. The broker may or may not be taking a good position on behalf of its clients. Believe it or not, companies do dumb things (occasionally).
And if not the official closing on the last day of the year, what else would a broker use?

The broker DOES have to justify what they use for tax reporting when they report basis...and it has to be the same across thousands of clients. They aren't going to pull a number out of thin air.
Another complication is that for spreads, individual legs of the spread can get reported at prices very far from a reasonable market value. For example say you buy synthetic stock (long call short put) as part of your box spread. There will be a very liquid market for the spread, because it's just based off of SPX which is very liquid, plus easy to determine adjustments for interest rates and dividends. Say you compute the fair value to be $9.99 and manage to buy the spread for $10.00.

First we need to think about how each individual leg of the spread gets reported. If the bid/ask spreads of the underlying call and put are small then there is not much room for error because they must get reported within the bid/ask spread, but for an expiration 4 years from now they might be wide. So maybe the Call has a bid/ask of $14/16 and the Put has a bid/ask of $4/6. How do you report each leg of the spread? In my experience it can be arbitrary - maybe they will report that you bought the call for $15.50 and sold the put for $5.50 - but lets just say for simplicity the exchange uses the midpoint of the bid/ask, so you buy the call for $15 and sell the put for $5.

You do this in the afternoon on 12/31 so you think your broker will show no gain or loss for tax purposes. But then at the very end of the day someone comes in places a market order for the Call, buying it for $16. If your broker uses the last trade price, it will show that your spread is worth $11. That's clearly unfair since your position is a synthetic long and SPX didn't moved 10% in the last afternoon of the year, but it's what you get from just using the last trading price.

Again, no idea how the brokers actually handle this, but there's not a perfect way to do it and there's definitely a risk of weird things happening.
I know the brokers use proprietary algorithms with interpolation techniques to determine the "mark price" of options; it's not the last trade price; they do it probably to avoid exactly that - to avoid the mark price fluctuating based on single trades. But I don't know how exactly the year-end values are determined, and if the year-end values for tax purposes are the same as mark prices. It doesn't really matter in most cases, as the total over the years doesn't change; it might matter if you are trying to do accurate tax planning perhaps to stay within a certain tax bracket; in which case you probably shouldn't use futures or options.
ggtt
Posts: 23
Joined: Thu Jan 19, 2023 11:32 am

Re: Let's Talk SPX Box Spreads

Post by ggtt »

kxl19 wrote: Wed May 17, 2023 10:52 am I'm curious if there was a hypothetical arbitrage opportunity a few weeks ago when the box spread yield was lower than a similar maturity treasury? (assumes no US default)

Example: Box spread June maturity was about 5.3%
Mid-June Treasury: yielding 5.5%+

Seems like the arbitrage opportunity gets bigger if you consider "after-tax" cost of box spreads (assuming the losses can offset existing gains) and similar after-tax yield of a treasury (interest tax exempt)

Anything I'm missing from this logic?
Counterparty risk of the treasuries is arguably higher due to the debt limit
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Let's Talk SPX Box Spreads

Post by comeinvest »

kxl19 wrote: Wed May 17, 2023 10:52 am I'm curious if there was a hypothetical arbitrage opportunity a few weeks ago when the box spread yield was lower than a similar maturity treasury? (assumes no US default)

Example: Box spread June maturity was about 5.3%
Mid-June Treasury: yielding 5.5%+

Seems like the arbitrage opportunity gets bigger if you consider "after-tax" cost of box spreads (assuming the losses can offset existing gains) and similar after-tax yield of a treasury (interest tax exempt)

Anything I'm missing from this logic?
Tax on treasury is ordinary income tax rate, right? Where is the tax arbitrage?
kxl19
Posts: 248
Joined: Wed Mar 13, 2013 12:41 am

Re: Let's Talk SPX Box Spreads

Post by kxl19 »

Tax on treasury is ordinary income tax rate, right? Where is the tax arbitrage?
Back of envelope -

Box Spread Yield 5.3%

After tax "borrowing cost" (assuming high tax bracket + offsets existing gap gains) approx 60% * 5.3% = 3%

Treasury - After tax yield (37% tax + NIIT) = .6* 5.5% = 3.3%

Treasury yield (3.3%) > borrowing cost (3%)
Kbg
Posts: 355
Joined: Thu Mar 23, 2017 11:33 am

Re: Let's Talk SPX Box Spreads

Post by Kbg »

comeinvest wrote: Wed May 17, 2023 11:10 am I know the brokers use proprietary algorithms with interpolation techniques to determine the "mark price" of options; it's not the last trade price; they do it probably to avoid exactly that - to avoid the mark price fluctuating based on single trades. But I don't know how exactly the year-end values are determined, and if the year-end values for tax purposes are the same as mark prices. It doesn't really matter in most cases, as the total over the years doesn't change; it might matter if you are trying to do accurate tax planning perhaps to stay within a certain tax bracket; in which case you probably shouldn't use futures or options.
This is kind of my thought as well...if some off marked options are going to blow your entire year's tax planning, yeah...probably ought to take a pass. And, in the larger picture of things...it will become accurately priced if held to maturity at which point you will will be made whole (or pay).
tj-longterm
Posts: 592
Joined: Tue Apr 10, 2007 10:10 am

Re: Let's Talk SPX Box Spreads

Post by tj-longterm »

kxl19 wrote: Wed May 17, 2023 11:53 am
Tax on treasury is ordinary income tax rate, right? Where is the tax arbitrage?
Back of envelope -

Box Spread Yield 5.3%

After tax "borrowing cost" (assuming high tax bracket + offsets existing gap gains) approx 60% * 5.3% = 3%

Treasury - After tax yield (37% tax + NIIT) = .6* 5.5% = 3.3%

Treasury yield (3.3%) > borrowing cost (3%)
Assuming the after-tax yields actually work out, I still wouldn't call it an arbitrage opportunity. Arbitrage, properly, means risk-free and there's a reason the market has priced the yields like this. There is actual risk involved, even if you personally think the risk is low.
tj-longterm
Posts: 592
Joined: Tue Apr 10, 2007 10:10 am

Re: Let's Talk SPX Box Spreads

Post by tj-longterm »

Kbg wrote: Wed May 17, 2023 1:44 pm
comeinvest wrote: Wed May 17, 2023 11:10 am I know the brokers use proprietary algorithms with interpolation techniques to determine the "mark price" of options; it's not the last trade price; they do it probably to avoid exactly that - to avoid the mark price fluctuating based on single trades. But I don't know how exactly the year-end values are determined, and if the year-end values for tax purposes are the same as mark prices. It doesn't really matter in most cases, as the total over the years doesn't change; it might matter if you are trying to do accurate tax planning perhaps to stay within a certain tax bracket; in which case you probably shouldn't use futures or options.
This is kind of my thought as well...if some off marked options are going to blow your entire year's tax planning, yeah...probably ought to take a pass. And, in the larger picture of things...it will become accurately priced if held to maturity at which point you will will be made whole (or pay).
If you sell a million of box spreads for whatever reason, and an off mark causes you to add $300,000 in phantom 60/40 capital gains income, it seems like it could be potentially painful short term, even if it eventually gets reversed. It's not clear to me how likely this actually is.
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Let's Talk SPX Box Spreads

Post by comeinvest »

tj-longterm wrote: Wed May 17, 2023 3:37 pm
Kbg wrote: Wed May 17, 2023 1:44 pm
comeinvest wrote: Wed May 17, 2023 11:10 am I know the brokers use proprietary algorithms with interpolation techniques to determine the "mark price" of options; it's not the last trade price; they do it probably to avoid exactly that - to avoid the mark price fluctuating based on single trades. But I don't know how exactly the year-end values are determined, and if the year-end values for tax purposes are the same as mark prices. It doesn't really matter in most cases, as the total over the years doesn't change; it might matter if you are trying to do accurate tax planning perhaps to stay within a certain tax bracket; in which case you probably shouldn't use futures or options.
This is kind of my thought as well...if some off marked options are going to blow your entire year's tax planning, yeah...probably ought to take a pass. And, in the larger picture of things...it will become accurately priced if held to maturity at which point you will will be made whole (or pay).
If you sell a million of box spreads for whatever reason, and an off mark causes you to add $300,000 in phantom 60/40 capital gains income, it seems like it could be potentially painful short term, even if it eventually gets reversed. It's not clear to me how likely this actually is.
I have many large box spread positions, some of them with strike prices that are very illiquid. I monitor the values daily in TWS at Interactive Brokers. I never saw any major irregularity with the market values or in the daily P&L column. I think we had some discussions earlier in this thread whether an off mark can lead to a margin liquidation, which might be more damaging than a wrong year-end valuation. There are some vague and anecdotal stories of margin calls in accounts that seemed to have more than one issue with leverage, but I have found no trustworthy report on the internet about a margin liquidation due to an off mark liquidation. I see the point however; $300k at the high tax bracket for that number would not be reversed by -$300k the following year at a low or partially at no tax bracket.
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Let's Talk SPX Box Spreads

Post by comeinvest »

tj-longterm wrote: Wed May 17, 2023 3:18 pm
kxl19 wrote: Wed May 17, 2023 11:53 am
Tax on treasury is ordinary income tax rate, right? Where is the tax arbitrage?
Back of envelope -

Box Spread Yield 5.3%

After tax "borrowing cost" (assuming high tax bracket + offsets existing gap gains) approx 60% * 5.3% = 3%

Treasury - After tax yield (37% tax + NIIT) = .6* 5.5% = 3.3%

Treasury yield (3.3%) > borrowing cost (3%)
Assuming the after-tax yields actually work out, I still wouldn't call it an arbitrage opportunity. Arbitrage, properly, means risk-free and there's a reason the market has priced the yields like this. There is actual risk involved, even if you personally think the risk is low.
Many refer with "arbitrage" to "semi-arbitrage" i.e. a low risk play with likely positive return. I think in many cases a semi-arbitrage could be decomposed into a proper arbitrage and a very unlikely tail risk event, or you can view it as a position with very low combined volatility.

But more important than terminology is probably the question if the position would lie on the efficient frontier within a portfolio, if you have an uncorrelated and low risk, but low expected return position.
Kbg
Posts: 355
Joined: Thu Mar 23, 2017 11:33 am

Re: Let's Talk SPX Box Spreads

Post by Kbg »

comeinvest wrote: Wed May 17, 2023 7:07 pm I have many large box spread positions, some of them with strike prices that are very illiquid. I monitor the values daily in TWS at Interactive Brokers. I never saw any major irregularity with the market values or in the daily P&L column. I think we had some discussions earlier in this thread whether an off mark can lead to a margin liquidation, which might be more damaging than a wrong year-end valuation. There are some vague and anecdotal stories of margin calls in accounts that seemed to have more than one issue with leverage, but I have found no trustworthy report on the internet about a margin liquidation due to an off mark liquidation. I see the point however; $300k at the high tax bracket for that number would not be reversed by -$300k the following year at a low or partially at no tax bracket.
Don't mean to be a pollyanna about this stuff...but there are much larger fish in this pond than us and I highly doubt, given what is at stake for them EOY reporting wise, that the final option price of the year is going to be completely wackadoo.

It appears there are mechanisms that assist with keeping pricing in alignment...

https://www.cboe.com/us/options/market_ ... ng_prices/

And there are definitely many options that expire on the last trading day of the year
tj-longterm
Posts: 592
Joined: Tue Apr 10, 2007 10:10 am

Re: Let's Talk SPX Box Spreads

Post by tj-longterm »

I think you're all right that probably the year end reporting is reasonable. I haven't noticed the marks ever that far off during trading hours, only after close when the 5000 PUT frequently is reported to be worth a lot more. It's something on my mind.
I see the point however; $300k at the high tax bracket for that number would not be reversed by -$300k the following year at a low or partially at no tax bracket.
I think this isn't quite right. You can go back and amend prior year tax returns (up to 3 years) and offset prior year 1256 gains with current 1256 losses, so you would presumably fix your tax bracket from the previous year as well. But, you'd be stuck carrying the tax payment until you're able to amend the return.

Again, I agree that it seems like this is probably not an issue. Even the Dec '27 options seems liquid enough that the marks during market hours don't seem to swing wildly, at least at TD Ameritrade.
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Let's Talk SPX Box Spreads

Post by comeinvest »

tj-longterm wrote: Fri May 19, 2023 12:41 pm I think you're all right that probably the year end reporting is reasonable. I haven't noticed the marks ever that far off during trading hours, only after close when the 5000 PUT frequently is reported to be worth a lot more. It's something on my mind.
I see the point however; $300k at the high tax bracket for that number would not be reversed by -$300k the following year at a low or partially at no tax bracket.
I think this isn't quite right. You can go back and amend prior year tax returns (up to 3 years) and offset prior year 1256 gains with current 1256 losses, so you would presumably fix your tax bracket from the previous year as well. But, you'd be stuck carrying the tax payment until you're able to amend the return.

Again, I agree that it seems like this is probably not an issue. Even the Dec '27 options seems liquid enough that the marks during market hours don't seem to swing wildly, at least at TD Ameritrade.
That's correct, I forgot that you can carry back 1256 gains. But yes, the whole issue is probably not an issue.
richardm
Posts: 49
Joined: Mon Nov 20, 2017 6:42 pm

Re: Let's Talk SPX Box Spreads

Post by richardm »

FWIW I've carried boxes across a new year on TD Ameritrade and indeed I've had wacky marks generate phantom tax losses/gains. In my case it was something like $600 on perhaps $60-80k in boxes so the tax consequence was maybe $100-150.

Going forward I'm tempted to close ALL of my 1256 -- not just the boxes -- in the last trading days of December just to make the accounting easier.
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Let's Talk SPX Box Spreads

Post by comeinvest »

adamhg wrote: Thu May 04, 2023 2:15 pm
comeinvest wrote: Thu May 04, 2023 10:50 am I got a fill for a sell box order today, $200k, 06/30 expiration, at 5.23% implied yield, when the T-bill BID yield was ca. 5.3% at that moment. What's happening? Any explanation? It looks like the spread to treasuries was already lower than normal lately.
Risk of default from debt ceiling? I'm in the process of doing the same thing, moving my gov/treasury bond allocations to box spreads
I suspect the APY calc on boxtrades.com might be a little inaccurate:
The chart shows rates above 7% for the June expiration, which is unlikely correct when treasuries yield about 5%.
The calculator on the site shows a rate of 6.799% for a box spread cost of 998.5. It shows 8 "days to expiry".
With my spreadsheet formula "(L378/J378)^(365/M378) - 1" where M378 are the number of days, I get 7.09% assuming 07/07 start date and 07/15 end date, 6.28% assuming 07/16 end date, and 5.11% assuming 07/18 end date. Today is 07/07.
I know the terminology around "expiration", "options settlement" and "settled cash in account" is confusing, but regardless terminology, I think we agree that the day when "settled cash is in account" is the economically relevant day.
I think the 06/15 SPX options end trading on 06/15; the settlement price is determined based on the market on 06/16; and the cash settlement day is 06/19. So there are 11 days between settled cash (position initiated) and settled cash (position closed): From 06/08 to 06/19.
Please correct me if I'm wrong. But I think the 7.x% are not accurate for any meaningful intents or purposes.

Image
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Let's Talk SPX Box Spreads

Post by comeinvest »

I got an APY of 5.63% 5.15% (per my APY formula, paid 998.35 ignoring commissions) today for "buying" a 06/15 SPX 6000-7000 bear put (should trade at the same price as a $100k box).
I bought back this SPX spread "loan" that I initiated a while ago.
Ca. 5.5% 5.05% after commissions.
T-bills show a bid yield of ca. 5.05% and an ask yield of ca. 5%.
1 mo. Term SOFR is 5.13%.
At IB cash currently pays 4.580% (BM - 0.5%). Most broker money market funds currently pay ca. 4.9%.
I think this is a fantastic execution, isn't it?

Surprisingly, a "buy" order for a 3000-4000 June 30 box spread didn't fill at 996.5, which would have been an APY of 5.25% based on my APY formula. Although June 29 expiration treasuries trade at almost exactly the same yield as June 15 treasuries.

EDIT: I forgot the holiday.
Last edited by comeinvest on Thu Jun 08, 2023 1:17 pm, edited 1 time in total.
adamhg
Posts: 218
Joined: Sat Apr 10, 2021 8:40 am

Re: Let's Talk SPX Box Spreads

Post by adamhg »

comeinvest wrote: Wed Jun 07, 2023 4:40 pm I suspect the APY calc on boxtrades.com might be a little inaccurate:
The chart shows rates above 7% for the June expiration, which is unlikely correct when treasuries yield about 5%.
The calculator on the site shows a rate of 6.799% for a box spread cost of 998.5. It shows 8 "days to expiry".
With my spreadsheet formula "(L378/J378)^(365/M378) - 1" where M378 are the number of days, I get 7.09% assuming 07/07 start date and 07/15 end date, 6.28% assuming 07/16 end date, and 5.11% assuming 07/18 end date. Today is 07/07.
I know the terminology around "expiration", "options settlement" and "settled cash in account" is confusing, but regardless terminology, I think we agree that the day when "settled cash is in account" is the economically relevant day.
I think the 06/15 SPX options end trading on 06/15; the settlement price is determined based on the market on 06/16; and the cash settlement day is 06/19. So there are 11 days between settled cash (position initiated) and settled cash (position closed): From 06/08 to 06/19.
Please correct me if I'm wrong. But I think the 7.x% are not accurate for any meaningful intents or purposes.
You're right that its not taking any settlement dates into account. I can do the weekend easily enough don't want to commit to keeping holiday schedules updated. The other problem I have with it is that you're only considering it from the pov of somebody already in the position. If you were looking to enter the position, wouldn't you need to account for the settlement time for opening the trade? I've always figured using the expiration date was good enough a compromise. It's only when DTEs go lower that it becomes this problematic.

Maybe a better comprise would be to allow people to change the DTE so they can set whatever time frame they want to model for themselves?
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Let's Talk SPX Box Spreads

Post by comeinvest »

adamhg wrote: Thu Jun 08, 2023 12:26 am
comeinvest wrote: Wed Jun 07, 2023 4:40 pm I suspect the APY calc on boxtrades.com might be a little inaccurate:
The chart shows rates above 7% for the June expiration, which is unlikely correct when treasuries yield about 5%.
The calculator on the site shows a rate of 6.799% for a box spread cost of 998.5. It shows 8 "days to expiry".
With my spreadsheet formula "(L378/J378)^(365/M378) - 1" where M378 are the number of days, I get 7.09% assuming 07/07 start date and 07/15 end date, 6.28% assuming 07/16 end date, and 5.11% assuming 07/18 end date. Today is 07/07.
I know the terminology around "expiration", "options settlement" and "settled cash in account" is confusing, but regardless terminology, I think we agree that the day when "settled cash is in account" is the economically relevant day.
I think the 06/15 SPX options end trading on 06/15; the settlement price is determined based on the market on 06/16; and the cash settlement day is 06/19. So there are 11 days between settled cash (position initiated) and settled cash (position closed): From 06/08 to 06/19.
Please correct me if I'm wrong. But I think the 7.x% are not accurate for any meaningful intents or purposes.
You're right that its not taking any settlement dates into account. I can do the weekend easily enough don't want to commit to keeping holiday schedules updated. The other problem I have with it is that you're only considering it from the pov of somebody already in the position. If you were looking to enter the position, wouldn't you need to account for the settlement time for opening the trade? I've always figured using the expiration date was good enough a compromise. It's only when DTEs go lower that it becomes this problematic.

Maybe a better comprise would be to allow people to change the DTE so they can set whatever time frame they want to model for themselves?
Yes I take into account the T+1 settlement day of opening the position. My example above was based on a trade on 06/07, which would have cash settlement dates of 06/08 (opening) and 06/19 (closing). The perspective with respect to a trade settling on T of an investor already holding the position on day T would not be interesting, because either opening or closing on T would settle on T+1. which means any action and investment decision on any day T would only concern the period between T+1 and the date of settled cash at expiration (for either interest paid or interest received), unless you have a time machine.
I personally use my own spreadsheet, and frankly I adjust manually for Friday trade days or Friday expirations, as I never had time to study LibreOffice. But I'm pretty confident that LibreOffice or also most programming languages have some standard function to calculate working days between two days, or to calculate the dates by adding a certain amount of working days (T+1). Probably there are also functions in most software for U.S. holidays.
Yes the further out the expiration the less the error. Also, if you let the user input the number of days, it's manual labor every single time (lol) and it also skews the charts and mentally lessens the confidence if you don't know the magnitude of the error, or if you don't adjust you will get data points in the charts on Fridays or before holidays that are out of the trend just because of the artifacts from the wrong T+1 calc. I know programming is tedious and thank you for your work so far!!!
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Let's Talk SPX Box Spreads

Post by comeinvest »

I got an APY of 5.83% 5.28% (per my APY formula, paid 998.5 ignoring commissions) today for "buying" a 06/15 SPX 6000-7000 bear put (should trade at the same price as a $100k box).
Ca. 5.7% 5.17% after commissions.
T-bills show a bid yield of ca. 4.96% and an ask yield of ca. 4.87%.
Overnight SOFR is at 5.05%; 1 mo. Term SOFR is 5.15% https://www.cmegroup.com/market-data/cm ... -sofr.html

EDIT: I forgot the holiday.
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Let's Talk SPX Box Spreads

Post by comeinvest »

adamhg wrote:
Can you please remind me what is time zone of the trade timestamps on boxtrades.com ? Thanks.
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Let's Talk SPX Box Spreads

Post by comeinvest »

@adamhg
Is boxtrades.com updating? Last recorded trades are from 06/23.
Also, the 06/23 trades of the 09/15 boxes show APYs of ca. 5.27%. That seems unusually low. Prevailing T-bill rates as well as SOFR rates were about the same, ca. 5.25%.
EDIT: I think something is wrong with the APYs on boxtrades.com. For example, the trade of the Sep 15 4000-5000 spread on 06/16 for 986.5 shows an APY of 5.31%, but according to my calc the APY is 5.6% (not adjusted for the holiday) or 5.67% (correctly adjusted for the holiday). Both trade date and expiration are on a Friday.
Last edited by comeinvest on Wed Jun 28, 2023 9:52 pm, edited 1 time in total.
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Let's Talk SPX Box Spreads

Post by comeinvest »

For whatever reason, 3 mo. Term SOFR rates are now below T-bill rates. Are there still any debt ceiling issues?
3 mo. Term SOFR per CME web site https://www.cmegroup.com/market-data/cm ... -sofr.html : 5.24%
T-bill estimated and interpolated from IB quotes: 5.3%
My $200k "sell" Sep 29 SPX box today: 5.58%

Image

Image

Image
adamhg
Posts: 218
Joined: Sat Apr 10, 2021 8:40 am

Re: Let's Talk SPX Box Spreads

Post by adamhg »

comeinvest wrote: Wed Jun 28, 2023 1:32 pm @adamhg
Is boxtrades.com updating? Last recorded trades are from 06/23.
Also, the 06/23 trades of the 09/15 boxes show APYs of ca. 5.27%. That seems unusually low. Prevailing T-bill rates as well as SOFR rates were about the same, ca. 5.25%.
EDIT: I think something is wrong with the APYs on boxtrades.com. For example, the trade of the Sep 15 4000-5000 spread on 06/16 for 986.5 shows an APY of 5.31%, but according to my calc the APY is 5.6% (not adjusted for the holiday) or 5.67% (correctly adjusted for the holiday). Both trade date and expiration are on a Friday.
Hey thanks for letting me know the site wasn't updated. I found the error and should be fixed going forward.

I looked into the calcs, and it looks like you're using the YTM calculation for zero-coupon bonds. Boxtrades.com uses the bond equivalent yield (BEY) which I think is more comparable to tbills and simple interest from margin rates on the short side. It's also the recommend rate calculation method for box spreads by the OCC.

Details and differences between the two are below

Code: Select all

trade          6/16/2023 6/16/2023
exp/settlement 9/15/2023 9/18/2023
dte            91        94       
spread         1000      1000     
cost           986.5     986.5    
                                  
bey            5.49%     5.31%    
ytm            5.60%     5.42%    
Also, It looks like you're using 91 dte (ignoring holidays) because you're counting the settlement time for entry, but I'm only accounting for settlement time for expiration with the assumption brokerages will require you to have the buying power available, or at least I haven't found a way to keep it earning in anything else.
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Let's Talk SPX Box Spreads

Post by comeinvest »

adamhg wrote: Sat Jul 08, 2023 11:13 pm
comeinvest wrote: Wed Jun 28, 2023 1:32 pm @adamhg
Is boxtrades.com updating? Last recorded trades are from 06/23.
Also, the 06/23 trades of the 09/15 boxes show APYs of ca. 5.27%. That seems unusually low. Prevailing T-bill rates as well as SOFR rates were about the same, ca. 5.25%.
EDIT: I think something is wrong with the APYs on boxtrades.com. For example, the trade of the Sep 15 4000-5000 spread on 06/16 for 986.5 shows an APY of 5.31%, but according to my calc the APY is 5.6% (not adjusted for the holiday) or 5.67% (correctly adjusted for the holiday). Both trade date and expiration are on a Friday.
Hey thanks for letting me know the site wasn't updated. I found the error and should be fixed going forward.

I looked into the calcs, and it looks like you're using the YTM calculation for zero-coupon bonds. Boxtrades.com uses the bond equivalent yield (BEY) which I think is more comparable to tbills and simple interest from margin rates on the short side. It's also the recommend rate calculation method for box spreads by the OCC.

Details and differences between the two are below

Code: Select all

trade          6/16/2023 6/16/2023
exp/settlement 9/15/2023 9/18/2023
dte            91        94       
spread         1000      1000     
cost           986.5     986.5    
                                  
bey            5.49%     5.31%    
ytm            5.60%     5.42%    
Also, It looks like you're using 91 dte (ignoring holidays) because you're counting the settlement time for entry, but I'm only accounting for settlement time for expiration with the assumption brokerages will require you to have the buying power available, or at least I haven't found a way to keep it earning in anything else.
I'm not a banker so I'm not too familiar with the intricacies of interest rate calc conventions. But in my spreadsheet I use (copied from my spreadsheet): (L419/J419)^(365/M419) - 1. I think this may be called "continuously accrued credited interest", but don't quote me on this. Let's call this "method 1".

The OCC document uses 2 different methods for box spreads and for T-bills, with is nonsensical for comparison purposes in the first place. They use 999.40 (the initial amount paid or received) as the divisor for box spreads (let's call this "method 2"), but $1,000,000 (the value at maturity) as the divisor in their T-bill example, the resulting yield of which they refer to as "discount yield" (let's call this "method 3"). They also use 365 days per year for box spreads, and 360 days per year for T-bills.

The settlement days in our example are 06/20 (because of the holiday) and 09/18, a time period of exactly 90 interest-bearing days, not 91 and not 94. That's a fact, and not subject to opinion or convention. You are referring to the margin check at order entry, which is irrelevant to the interest earned or paid. When you buy the box, your cash will (by definition) earn interest in your brokerage account until the settlement date; when you sell the box, your cash received will (by definition) start earning interest in your brokerage account on the settlement date. If you need to transfer in money from somewhere else, and/or your broker doesn't pay interest, or other personal circumstances apply, then those have nothing to do with the transaction at hand. Cash can be used by most market participants to and from the settlement days, either for other transactions (e.g. stock purchase) based on their respective settlement days (T+2), or it will earn interest (or reduce interest payable, if you have an overall debit balance). Nothing to argue here.

The differences between the 3 methods should be small, because of the short maturity. Let's verify:

Method 1: (1000 / 986.5) ^ (365 / 90) - 1 -> 0.05667 -> 5.667%
Method 2: (13.5 / 986.5) * (365 / 90) -> 0.05550 -> 5.55%
Method 3: (13.5/1000) * (360 / 90) -> 0.054 -> 5.4%

The difference is at most 0.27% - not entirely meaningless.

With 91 days, the yield per method 1 would be about 5.603%.
With 94 days, the yield per method 1 would be about 5.419%.
But those yields are irrelevant because they are just plain wrong and meaningless for all economic intents and purposes.
We have to decide between the 3 yields in bold above. It would probably make sense to use the same method that is used by whatever instrument we are comparing to, as that would be our opportunity cost. I think for T-bills and SOFR futures that would be a discount rate; at least that's what is most visible based on their quotation convention; although I have not checked how the SOFR futures quotes are annualized. I'm not sure about LIBOR conventions, broker, checking, or savings accounts.

But I think you for sure want to correct your DTE in your calcs and adjust for weekends and holidays, to have any meaningful numbers for comparison.
adamhg
Posts: 218
Joined: Sat Apr 10, 2021 8:40 am

Re: Let's Talk SPX Box Spreads

Post by adamhg »

I think we're saying the same thing. Here's the same calculations I had with 90days:

Code: Select all

trade                6/16/2023 6/16/2023
exp/settlement       9/15/2023 9/18/2023
dte            90    91        94       
spread         1000  1000      1000     
cost           986.5 986.5     986.5    
                                        
bey            5.55% 5.49%     5.31%    
ytm            5.67% 5.60%     5.42%    

So what you have as "method 1" is what I had labeled as "ytm" as referenced here. "method 2" and what boxtrades.com is using is labeled as "bey" as referenced here.

It's been a decade since I took finance so I'm very rusty. But from my google-fu, "method 3" discount rate is "generally converted to bey".

So bey is how to compare against semi-annual bonds, which is how yield curves are described and what t-bills are converted into. It's the same calculation as simple interest, which is used to represent margin interest rates. So for both longs and shorts, it seems like the "bey" calculation is most commonly used to compare against various rates which is why I am showing that rate instead of ytm/continuously compounding rates.

I am working on holidays and weekends. But ultimately once everything's updated in the next few days, I'm committing the the site will show by default:
  • bond equivalent yield from opening settlement to closing settlement date accounting for weekends and holidays
Hope that helps clarify what I plan to show and why. I may end up adding "ytm/method 1" as an option people can toggle, but that will be a while away.
comeinvest
Posts: 2709
Joined: Mon Mar 12, 2012 6:57 pm

Re: Let's Talk SPX Box Spreads

Post by comeinvest »

adamhg wrote: Tue Jul 11, 2023 5:15 pm
Thanks adamhg. Thanks for adjusting for weekends and holidays.
Regarding the interest calc methods, I still have to wrap my head around them.
LearnMarket
Posts: 17
Joined: Tue Jun 16, 2020 4:53 pm

Re: Let's Talk SPX Box Spreads

Post by LearnMarket »

Hi All,
I am testing waters. I am trying to go Long on box trade. But I am not getting a fill. I have tried several times. I do not see volume on some of the legs. So the trade does not happen. It is not because the LIMIT price. Is this common? or Am I missing something? Any tips on how to get a fill?

For example.
$SPX
Buy 1 SPX 19JUL24 4000 CALL
Sell 1 SPX 19JUL24 4000 PUT
Sell 1 SPX 19JUL24 4050 CALL
Buy 1 SPX 19JUL24 4050 PUT
User avatar
Topic Author
GeraniumLover
Posts: 560
Joined: Mon Feb 22, 2016 2:39 pm

Re: Let's Talk SPX Box Spreads

Post by GeraniumLover »

LearnMarket wrote: Wed Jul 19, 2023 3:12 pm Hi All,
I am testing waters. I am trying to go Long on box trade. But I am not getting a fill. I have tried several times. I do not see volume on some of the legs. So the trade does not happen. It is not because the LIMIT price. Is this common? or Am I missing something? Any tips on how to get a fill?

For example.
$SPX
Buy 1 SPX 19JUL24 4000 CALL
Sell 1 SPX 19JUL24 4000 PUT
Sell 1 SPX 19JUL24 4050 CALL
Buy 1 SPX 19JUL24 4050 PUT
Did you try a 3950/4000 spread?
Post Reply