April 17, 2024 – People sometimes ask me for a good and safe place to “park” their money for a short period. CDs, high-yield savings, and money market accounts would be the obvious answers. When looking for safe, short-term investments, options are probably the last thing on your mind. Options have the aura of complicated and highly speculative investments. However, sophisticated investors can structure options trades to make them (almost) as safe as CDs but with more flexibility and higher after-tax income, thanks to a Box Spread trade.
You can implement this trade by hand, and I will go through the mechanics. You can also buy an ETF, though with some small drawbacks. Let’s take a look…
Avid readers of the ERN blog will know that today’s post is an extension of an older post where I proposed box spreads to borrow against your portfolio. But we can certainly reverse the box trade and use it to lend instead of borrow funds. So, let me explain again what a box spread is and how you can trade one at Interactive Brokers, a site I heavily use for all my options trading needs.
What is a Box Spread?
As I detailed in my post from a while ago, a box spread consists of four options positions at two separate strikes: two puts and two calls. At each strike, there is one put and one call. Because the payoff profile of this set of four options is strictly a straight line, equal to the spread between the two strikes, we’ve essentially constructed a synthetic zero coupon bond that pays off a fixed amount regardless of the underlying at the time of expiration.
Here’s a concrete example: We buy the S&P 500 index Call with a 4000 strike and the Put with a 5000 strike. And also sell the 5000 Call and the 4000 Put. In the chart below, I display the P&L from the four underlying positions as well as the total: The four lines always add up to 1000 points no matter where the index lands on the expiration date.
In this example, it might be easier to see the mechanics of the box spread when displaying the same numbers in a table:
- If the underlying is below 4000, the calls expire worthless, and we receive the put spread of 1000.
- If the underlying is above 5000, the puts expire worthless, and we receive the call spread of 1000.
- Between 4000 and 5000 points, the short call and short put expire worthless, while the payoffs from the two long options exactly sum to 1000.
Also, notice that the spread is quoted in index points, while the CBOE index options each have a multiplier of 100x, so the total payout in December is $100,000.
Warning: You should only do box trades with “European Options,” e.g., CBOE SPX options, futures options, etc., that cannot be executed before expiration. Never do a box trade with “American Options,” e.g., options on individual U.S. stocks! Early execution of one of the short option legs before expiration would destroy the constant P&L line in the chart above! People who are confused about the difference between European and American options can lose a lot of money.
How do we trade this combination of four options? Do we enter four separate market orders for the four different legs? Better not! You’ll likely encounter large bid/ask spreads, especially for the deep-in-the-money options. Interactive Brokers offers a combination trade that’s much more efficient, ensuring tighter spreads and that all four legs trade simultaneously.
Here’s the implementation in detail:
Box Spread trading at Interactive Brokers: A step-by-step guide
1: Start a new watchlist and click the menu. Under Trading Tools, click “Strategy Builder:”
2: This opens a new window. In the field at the top left, I use “SPX index” and make sure to keep the other pulldown menu at “PUT/CALLs (Side by Side).”
3: The first four expirations only go to July 18, 2024. But when I click on “MORE,” I can find the December 19, 2024 expiration in the pulldown many. Click that:
4: This brings up the entire list of strikes in the middle column. Calls are on the left, and Puts are on the right. Imagine we like to generate a 1000-point spread (for a $100,000 payout). Let’s look for between 4000 to 5000 points. I click the ASK price for the 4000 Call, the Bid Price for the 4000 Put, the BID price for the 5000 Call, and the ASK price for the 5000 Put. Each click generates a “leg” of this strategy, and we end up with the four legs listed at the bottom of the window. IB is smart enough to recognize this combination as the “DEC19 4000-5000 Box,” which is a great error check. We can also quickly confirm that this 1000-point payoff costs you about Bid=962.50 to Ask=967.30 points. So, the $100k, effective zero-coupon bond would cost us about $96,500 if we can buy this spread at around the midpoint. That means we’d spend $96,500 today to receive $100,000 in December, i.e., a guaranteed gain of $3,500.
We can click “Add to watchlist” at the bottom to export this to our Box watchlist. We should also iterate this process a few times and add a few more 1000-point spreads. Click “Clear all legs” and populate a new 1000-point spread, like 4500-5500 and 5000-6000. I recommend using strikes close to the current SPX level and round strikes. There is probably very little activity in the 6400-7400 or 1000-2000 box spreads!
5: On that watchlist, we can now track the quotes of the three spreads. The best bid was 962.5 for the 4000-5000 spread, and the best ask was 963.30 again in the 4000-5000 spread. I took this screenshot around noon PDT on 4/16.
6: I right-click on the 4000-5000 spread and click “Buy” (=lend) this box spread and set the price to the midpoint of 962.90. I can also right-click the order line and select “Check Margin/Performance Profile” to perform one last error check; see the screenshot below. This trade triggers a $96,290 debit, and the commission is around $6. I can double-check the four underlying legs again and that the trade does not impact my margin constraints (except for the small commission).
I didn’t submit the buy order in this case because I already have several long box spreads expiring on 12/19/2024, but this is how I’ve previously entered my trades.
How much money can I earn with a Box Spread?
In the example above, trading the 12/19/2024 box spread on 4/15/2024, let’s look at what kind of implicit interest rate I would have earned. Let’s take the best bid and ask prices as well as the midpoint of 964.10 to gauge how different prices would yield different IRRs. I use the quotes 962.50 (highest bid), 963.30 (lowest ask), as well as the midpoint of 962.90 and a commission of $6.00. We’d have to spend between $96,256 and $96,336 to guarantee a future $100,000 payout. That translates into an IRR between 5.65% and 5.78%. You could get the 5.65% immediately if you buy at the ask, but chances are that you may get a better yield if you’re patient and put in your bid at the midpoint.
What is a good price to fill this order? My experience is that you target the risk-free Treasury yield of the same duration plus about a spread of 0.30-0.40 percentage points. So, with the 1-year Treasury yield at 5.17% on that day and the 6-month T-Bill at 5.36%, interpolate that to maybe 5.30% for the 8-month yield, plus 0.35%-0.40%, and you get right to 5.65%. So, even at the ask price, that’s a very competitive yield, but you may get a fill at a slightly lower price and higher yield as well, maybe even the midpoint.
In any case, why all this effort for 5.7%? Why not just get a CD? Fidelity offered CDs at around 5.25%-5.35% on the same day for 6mo and 9mo CDs, respectively:
You might get 5.50% or even 5.70% by shopping around. Why all this hubbub for just a few basis points of extra yield? This brings me to the next point.
The Box Spread Tax Advantage
Gains from SPX index options enjoy preferential treatment on your tax return. Even though you hold your December Box Spread for only about eight months, the gains are treated as Section 1256 income, taxed as 60% long-term and 40% short-term gains – reported on IRS form 6781 – regardless of the holding period. So, let’s look at how the three different box spread IRRs translate into after-tax rates of returns and how those returns compare to CDs, where your income is subject to the ordinary income tax rate. In the table below, I calculate the after-tax return of the box spreads in four different tax regimes:
- A low-tax regime, where our investor is in the 0% tax bracket for long-term gains and 12% for ordinary income. This would apply to early retirees with an annual income below about $123,000. For example, married couples filing jointly: $29,200 standard deduction plus the top of the 12% bracket $94,300 = $123,500. Or $123,250 when using the top of the 0% long-term capital gains bracket.
- In an intermediate tax regime, you’re in the next tax bracket and pay 15% and 22% for long-term gains and ordinary income, respectively. This could be a FatFIRE retiree with income in the low six figures but already in the 22% federal bracket.
- You are in the top federal tax bracket of 37% and 20% for ordinary income and long-term gains, respectively. You also owe the 3.8% additional marginal tax for Obamacare.
- Let’s not forget about state taxes. The fourth regime has the highest federal bracket plus an additional 10% for state taxes.
To calculate the after-tax return of the box spread, I take the weighted tax rate (60% LT gains plus 40% ordinary income) and subtract that from the box spread yield. For comparison, I compute the “CD Equivalent,” i.e., the rate you’d need to fetch in a CD to match the box spread’s after-tax return. That’s easy to calculate; the box spread after-tax return is divided by one minus the ordinary income tax rate. For example, I’d need a 6.21% rate in a CD in the low-tax regime to match the 5.74% in the box spread. CD rates above 6% are not so easy to find right now!
The tax advantage of the Box Spread is a little smaller in the intermediate federal bracket. That’s because the S.1256 marginal tax rate is not too far below the ordinary tax rate (17.8% vs. 22%). However, with the higher tax brackets, the box spread becomes very attractive again relative to a CD. The ordinary income tax rate is 10.2 percentage points higher than the S.1256 contract marginal rate. Also, even when adding a state tax across the board and leaving the spread the same, the CD equivalent still becomes larger (dividing by one minus the ordinary marginal tax rate, we divide by a smaller number!). In the highest tax regime, you’d need to earn 6.93% in the CD to match the 5.74% in the Box trade. High earners in high-income-tax states will benefit most from the box trade!
How do I close the Box Spread?
There are multiple routes:
- Wait until the expiration, when all options expire, and you get your money back. There is no commission to pay.
- Close the box spread early by simply reversing the trades I explained above.
- What happens if your original box spread now has strikes far away from today’s index? Imagine you buy a 4000-5000 box spread, but subsequently, that spread now has wide bid-ask spreads. Can you sell a 5000-6000 box spread if its prices seem more attractive? Certainly. You will then hold multiple a collection of options at three different strikes, 4000, 5000, and 6000, that precisely offset each other and will expire with a $0 payoff at the expiration; see the table below:
And finally, you could also partially undo the $100k notional box spread. For example, if you have a long $100k notional box spread and need “only” $50k, you could borrow through a 4500-5000 box spread (or a 5000-5500 spread, or 5500-6000, whatever offers you the best execution). This move will partially liquidate the existing spread. You’re really only limited by transaction costs and B/A spreads. To save money, of course, it’s always best to hold the box spreads to maturity, which is what I’ve been doing so far.
Too lazy to trade the spread yourself? Try a Box Spread ETF!
If you’re overwhelmed with trading this yourself, do not despair. There’s an ETF for that. The Alpha Architect 1-3 Month Box ETF (BOXX) has been trading since late 2022. The advantage is that you can trade this in small lots and commission-free at Fidelity or most other retail brokerage firms, like Vanguard, Robin Hood, etc.
What kind of return can I expect from the BOXX ETF? It’s not so straightforward. The ETF does not pay out its box spread income but rather accumulates the gains. The advantage is that if you hold the ETF for more than a year, you can structure a CD-like return with the tax treatment of long-term gains. Sweet; that seems even better than building your own box spread, but you also pay the ETF expense ratio (0.19%). Also, the current BOXX yield seems a bit low. Since the ETF doesn’t pay any monthly dividends, we have to gauge the current yield from the price chart. That’s no easy task because the price chart for this ETF is quite bumpy; see the chart below (top panel). I tried two different routes to measure the yield:
- Calculate a 4-week moving average chart to smooth out some of the bumps. Then, take the 4-week return of that smoothed line and annualize it. That’s the blue line in the bottom chart. Currently, the annualized yield is about 5.12%.
- Calculate a Hodrick-Prescott Filter smoothed line (the red line in the top chart) and then calculate its 4-week return and annualize that, which I plotted in the green line in the bottom chart. The HP-filtered line is a bit better at eliminating the noise. The BOXX ETF price time series seems to have a slope consistent with a 5.20% annualized return.
Next, let’s perform the same calculations as before, i.e., calculate the BOXX ETF after-tax returns, assuming you can hold this fund for more than a year to structure your interest income as 100% long-term gains. To calculate the CD equivalent, we again divide the BOXX after-tax income by one minus the ordinary income tax rate; see the table below. If the BOXX yield is 5.20%, you’d still do better with the homemade Box Spread trade in the low and intermediate tax brackets. In contrast, the higher income tax brackets would benefit slightly from the BOXX ETF.
Some words of caution
Just for the record, there are some drawbacks of trading box spreads:
- More complexity: Sophisticated investors and financial advisers might benefit from this trade, but many ordinary investors are unwilling to trade options. I fully sympathize with that. If you invest a few thousand dollars, the tax advantage may not be worth the extra complexity. For larger amounts, though, six figures and up, you’re leaving a big chunk of money on the table by ignoring the box spread route.
- Credit risk: while your CD is insured through the FDIC up to $250k per bank, your box spread is only as credit-worthy as the options exchange. Even for very creditworthy players like the CBOE, you’d probably need to assume that about 0.3 percentage points of any excess yield might just be a credit spread, not free excess yield.
- Tax uncertainty: While the self-made box spread will likely always enjoy the 60/40 long-term/short-term tax treatment, some folks are concerned that the IRS may eventually crack down on the BOXX ETF, flaunting the ETF retained gains tax loophole. Before too long, the IRS may force the ETF provider to distribute the S.1256 gains regularly, i.e., monthly, like most other short-term fixed-income ETFs (e.g., SHV). Paying S.1256 on 5.2% income is still better than paying ordinary income tax on 5.2%, so BOXX would still have a tax advantage, albeit a smaller one. In that case, BOXX would fall behind the homemade box spread even for the high marginal tax households.
- Longer Horizon: To utilize the BOXX ETF tax advantage, you’d need to hold it for one year plus one day. Ideally, one would create a ladder of BOXX ETF tax lots so that we can always liquidate the shares that already qualify for long-term gains. If you have a specific cash flow need within one year, you’re better off using the homemade box spread.
- Illiquidity: Both box spread routes have some liquidity risk. The BOXX ETF has some volatility, and if you need your money back on a specific day, you may get a price a few cents below the trend line. Likewise, you may be unable to unwind the box spread quickly due to a potentially large bid-ask spread and nobody biting on your limit order. Your limit order might sit there for a few days until it goes through. Then again, some CDs also have the same problem (and potentially worse!) in the form of early redemption penalties.
Which Box trade is for you?
Using a side-by-side comparison, implementing the box trade yourself will likely beat the BOXX ETF. However, the BOXX ETF may work for smaller investors who like to dip their toes and make smaller investments and withdrawals. I trade both: 1) the homemade box trade at Interactive Brokers, where I trade all my other options. And 2) the BOXX ETF in our Fidelity brokerage account, which we use for cash management, i.e., where we have a few months’ worth of cash to pay our bills.
Conclusions
With the equity market sputtering a bit, uncertainty about inflation, interest rates, and the November election, some folks wonder if it’s wise to park some money in a safe asset for a while. With 5+% interest rates, that makes more sense than only a few years ago when interest rates were at rock bottom. I showed a few options with better yields than your average CD or money market fund, especially when you’re concerned about the after-tax yield, not just the headline number, where you run the risk of comparing apples and oranges, i.e., tax-advantaged box trades vs. low tax efficiency CDs and money market interest income. It’s not for everyone, but I’m just here to offer options, pun intended.
Re: Tax uncertainty
BOXX is using the ETF heartbeat trick to avoid distributions. I don’t see how IRS can force them to make distributions without changing the law and breaking all the other ETFs.
However there is another provision of tax law. If an instrument behaves like a bond, they can force you to treat it as bond income. As far as i can remember, this requires a special ruling. And this would break all long-box positions not just the ETF. So probably this is also unlikely and would apply forward not backward.
I think until IRS makes a move, using ETF has minimal risks and more adventageous than trading boxes yourself if you intend to hold more than 1 year.
If you intend to hold less than 1 year, then with manual boxes you get 60/40 tax treatment vs with the ETF you get 100% short term gains taxed.
Thanks for the explanation!
The risk of “breaking all ETFs” is one that some industry observers are worried about: if the BOXX ETFs gets big and attracts the ire of the IRS, it would spoil the tax advantage of all ETFs:
https://elmwealth.com/thinking-outside-the-boxx/
The second issue you raise, the IRS declaring BOXX a bond-like instrument, is the more worrisome for BOXX but the less worrisome for the industry.
I think calling a four-legged option strategy a “bond” would not hold up in any court.
They don’t have to call it a bond, just a “conversion transaction” as defined in IRC “§ 1258 Recharacterization of gain from certain financial transactions”.
“substantially all of the taxpayer’s expected return from which is attributable to the time value of the taxpayer’s net investment in such transaction, and…”
There’s more to the “and…” but the taxnotes article below makes a compelling argument that box spreads meet the requirements.
Note also IANAL but I believe much of this applies to individuals lending money with box spreads as well. BOXX may have attracted enough attention that the IRS is going to provide guidance on the taxation of lending and borrowing with box spreads. Of course, they aren’t exactly known for their speedy responses.
https://www.taxnotes.com/featured-analysis/tax-trap-inside-boxx/2024/03/08/7j8x0
Thanks for the info. I guess we should milk this BOXX and Box tax hack for as long as it lasts. BOXX currently has $2b AUM, so I don’t think that the IRS is too eager to move fast on a few million dollars of lost revenue.
The language in S. 1258 is clear. So there is a law that addresses investments like this that behave like a bond. Question is: would even the language of S.1258 hold up in court.
I read an article in Forbes specifically arguing that box spreads and BOXX are conversion transactions and shouldn’t get any special tax treatment. Granted, this is just one man’s opinion, but I personally agree with his argument.
https://www.forbes.com/sites/stevenrosenthal/2024/03/04/boxxs-tax-gimmick-violates-congress-rules-on-conversion-transactions/?sh=58d3a460291d
While I think long box spreads shouldn’t get the S.1256 tax treatment, the IRS currently has no way of identifying people trading them. Trades of individual S.1256 contracts aren’t reported on any tax forms, so the IRS would never know you’re trading long box spreads. They would only see the single P/L number from all aggregate S.1256 trades. But trades of the BOXX ETF would be known to the IRS on a 1099-B.
Yeah, I saw that article, too. Agree with your assessment: it’s hard for the IRS to enforce this anytime soon, at least for home-made box trades. The BOXX ETF gains would be known, but likely not the itemized gains by ETF. It’s my understanding that the IRS only knows the aggregate numbers for LT and ST gains/losses.
Thank your great articles in your blog.
I‘m a fan. Could you recommend some boxx etfs? Warm regards from Germany Hartmut
Hi Hartmut! The only ETF I know is the one mentioned in the post, Ticker=BOXX.
Viele Gruesse!
One wrinkle I’ve run into holding long boxes across the end of the year is related to the marked to market taxes on 1256 contracts. The options making up the spread are taxed based on the last traded price, not the mark price. Since some of these options are typically deep in the money the options may not trade frequently. This can cause the last traded prices and thus the tax liability to be quite different than you might expect based on the current marked price. I had spreads where this mismatch between last and mark prices completely erased the gains I made on my spreads during a year. This was actually bad for me in terms of overall taxes owed. Although I didn’t owe any taxes in the first year, now the taxes in the following year will make my overall taxes higher due to shifting where my income falls in the tax brackets in the second year. Also, broker specific, I’ve found that Schwab does not report unrealized gains using the last traded prices anywhere on their website, so my tax liability will actually be unknown until I get my 1099 which is past when estimated taxes are due. While last prices are available through their trading platform (TOS), those prices are not guaranteed to be the same as the values reported on a 1099. Obviously you can avoid this issue if you don’t hold boxes across calendar years, but if you trade out and back in to all your positions at the end of one year and start of the next, you’re paying extra commissions and dealing with slippage on order entry and exit multiple times.
Interactive Brokers (IB) handles this well. All mark to market is done smoothly, even though some of the far-ITM options have wide B/A spreads. I think they price the “combination” (which they call a “complex position”) as a bond-like entity and then assign prices to the 4 subcomponents.
If your broker causes this kind of tax uncertainty, that would be too bad. Maybe don’t do boxes or move elsewhere.
Hm. My case is similar to high federal + 10% state. But I think the appropriate benchmark is treasuries as they have no state income tax rather than CDs.
This seems to effectively eliminates the tax advantage of box spreads (50.8% in your example vs. 40.8% for treasuries vs. 40.6% for box spreads. )
At that point the benefit is just the additional yield you might get via the box spreads of, say, ~.4% and after taxes that is ~.25% or so for the hassle.
So perhaps a nice-to-have problem, but it seems to marginalize the benefit of box spreads.
Is that fair?
Mark, have you considered what happens in years when you incur both short-term and long-term capital losses? Any recognized gain from a box spread can be directly offset by these capital losses. This is a significant advantage—it means I’ve not only realized income from excess capital but also paid no taxes on that income. Also, many state tax returns start with AGI, thus lower AGI to start with, the better. I do hold treasuries just a a little as very smal emergency reserves; otherwise, I’m actively engaged in box trades. To me after I learned them are no-brianer.
Personally, I’ll consume any TLH regardless of the box spread strategy. So I wouldn’t bucket that under box-spread-benefits though perhaps for others that might be meaningful if they have no other way to consume the tax loss harvesting.
Would you agree with my high level read in the prior post though (for my or others with similar circumstances)?
Good point. Some folks sit on large piles of old capital losses that they are working off at $3,000 per year. With this approach you can accelerate the process.
If you have some Long & short term capital losses, I would rather apply the gains against those vs. paying the federal tax on the treasuries
Bingo! Peter got it. Mark not so much and that’s okay – Mark gets “gold star” award from Treasury!
Good point. Is this true for not just individual bonds (hard to trade) or also bond ETFs like IEF, TLT, etc.?
Scanned through the comments since it’s often where the best stuff is, and saw this conversation. Treasury ETFs can be state tax-free, usually partially. TFLO and USFR are mostly or entirely state tax-free. Fidelity publishes a list each year of its own funds and what % is state tax-free. It’s not easy to navigate this stuff and fill out a tax return properly, since 1099s don’t include this information and you have to manually find the broker publications and enter it properly into tax software. Turbotax does handle this fine if you know where to look.
True. I miss those details sometimes because I’m in WA state with no state income tax. If you’re in a high-tax state you may want to factor that in. It might get better after-tax results than even the S.1256 method.
When I thought FIRE couldn’t get more complicated, I come here to Big ERN. He always proves it can get more complex and mathy
Well, earning more after-tax income makes life easier, but you’re right, we can always make life more complex, haha!
Hi ERN
Great post as usual. I think there’s a small typo here: “I click the ASK price for the 4000 Call, the Bid Price for the 5000 Put, the BID price for the 5000 Call, and the ASK price for the 5000 Put.” You meant “the Bid Price for the 4000 Put” right instead of the 5000 Put?
Also on a separate note on your other post regarding trading SPX options, is there a tangible risk of stop losses being blown through and suffering a larger loss than expected in terms of market turmoil (for example you set a 2-week worth of premiums stop loss but the stop loss order doesn’t execute or executes at a price that is 1 month worth)? Or is that risk minimal in your opinion with a stop loss-market order (as opposed to a stop loss-limit order) or on SPX?
Yes, that was a typo. I corrected the section: “I click the ASK price for the 4000 Call, the Bid Price for the 4000 Put, the BID price for the 5000 Call, and the ASK price for the 5000 Put”
About the Stops: Yes, that’s a concern. Usually the strikes are far enough OTM, so their price rises slowly to the STP and then executes at the STP price plus 0.05 to 0.15.
Thanks ERN. How do you manage to set a box spread as “liquidate last” on IBKR? That ‘option’ is greyed out for me (or you don’t think it matters)?
Appreciate it – always just concerned with a flash crash and then getting liquidated at significantly worse prices.
I don’t think you can do that.
So, in the case of a margin call you might risk that your broker break up a box trade and liquidates only one of the legs. That’s quite dangerous. Always keep an extra margin cushion!
A long box spread is a section 1258 conversion transaction. If the IRS ever gets its act together it will go after this conversion transaction whose income consists solely on the time value of money, and tax it as interest.
To do that consistently for all options trades, wouldn’t they have to do things like calculate the interest rate contribution to the decay of a covered call so that they could separate out the interest from the capital gain/loss portion of the outcome?
Yeah, if you buy 3-year box spreads and then sell them after 2 years, then how much of the gain is interest vs. capital gains/losses (due to a duration effect)? Who calculates that? The broker? The investor? The IRS?
It’s a concern for the fans of boxes. Probably more so for the BOXX ETF. For investors in homemade boxes, it’s not clear how the reporting from your brokerage would updated to capture the boxes. You only report the net P/L from S.1256 contracts. Unless this is itemized into S.1258 vs. purely S.1258 contracts there is no way investors will be able to file this properly.
I guess to quantify the concern, we’d need to know whether a court ruling on the matter could be retroactive. Legislation is generally applicable from a point moving forward, but what if in December, 2024 a court ruled that box investors owed a bunch of taxes on interest received in 2024? The interest attributable to each BOXX investor throughout the year would be hard to calculate, and the tax forms would arrive as a surprise.
The flipside of such a ruling would be that box borrowers would be able to write off their expense as interest expense. The same court ruling that would make boxes into bonds would open up the strategy to all borrowers. Corporations might be interested in this new way of “borrowing” at close to the risk-free rate. Investors would find this strategy cheaper than their broker’s margin. With enough liquidity, people might mortgage their homes or finance their cars this way. So much for banks!
Yet, I don’t know if such a flood of demand for short boxes would raise the “interest” rate like it would for other types of debt. Either our options pricing models would have to be rewritten to account for the change in rules or option market liquidity would collapse as there would always be more people seeking to borrow at near the risk free rate than there would be to lend.
Yes, good point. Right now I’m still a net-borrower with boxes, though I have reinvested all my recent Preferred share redemptions back into long boxes. So, I might be able to write off 100% ordinary income instead of 60/40? Bring it on!
But other investors, box long-only, can hopefully milk this tax hack for a while longer.
I’ve experienced a strange number of early call assignments, but these are largely for ETFs. I guess this would *never?* happen for cash-settled short index options?
Seems like if one tried this with ETFs, an unforeseen assignment would quickly throw this neutral strategy into a large directional bet.
You would only use European options (no early assignment) as legs in this strategy. CBOE SPX options, futures options, etc. Never options on individual stocks.
The ERN comment above could be in bolded caps and still be understated.
https://www.marketwatch.com/story/trader-says-he-has-no-money-at-risk-then-promptly-loses-almost-2000-2019-01-22
Very good point. Never do this “no money at risk” with American options.
I added the warning in prominently and in bold big letters in my post!
Hi ERN,
Thanks for the great post and very helpful step-by-step guide.
I have a few questions about the detail:
1. In step 6, the price range is 962.50-963.30 with midpoint 962.90. but it says “This trade triggers a $92,900 debit, and the commission is around $6”. Why is the debit only $92,900 instead of 962.90×100=$96,290 ? Is this a typo ?
In the next section “How much money can I earn with a Box Spread”, it says “We’d have to spend between $96,256 and $96,336” which is consistent with the price range and $6 commission.
2. For this box trade example, does it mean that you need to have ~$96k uninvested cash in Interactive Brokerage account to be able to trade?
Thanks.
WG
1: Sorry. That was a typo. I corrected the text.
2: It would be ideal to have the ~$96k in cash. If you don’t and you have to borrow to buy the box then you may pay more on margin interest than you make with the box.
What are the tradeoffs in expiration dates that are closer in (ie. May 31, 2024) vs. farther out (December 2024)? Besides the transaction costs, is one better than the other? What about the spread – better to go $1000 spread, $500 spread or doesn’t it matter?
I’ve only done far ahead spreads so far: borrowed in Dec 2025, all the way to Dec 2028. And now lending in December 2024. But I also put the July-November 2024 1000-point spreads on my watchlist now. The B/A spreads are tight and the interest rate is very attractive. I might transition to near-term spreads due to that.
In your Interactive Brokers example, you buy a 4000 Call and a 5000 Put while selling a 5000 Call and a 4000 Put. However, the numbers are reversed in your text at the top of the post where you say “Here’s a concrete example: We buy the S&P 500 index Call with a 5000 strike and the Put with a 4000 strike. And also sell the 4000 Call and the 5000 Put. ”
Is that a typo? Do both variants work or does only the later (IB example) correct?
Good catch. Those were typos.
If you had entered the complex position trade exactly reversed in the Strategy Builder, you would have gotten the same but with a negative price. So to lend you’d sell at a negative price (for a debit in your account).
Thanks, Mr ERN to unselfishly share the valuable knowledge.
BOXX has low risk, good rate, tax advantage. What’s the point for people to buy bonds or BND in the portfolio? Or even leave money on savings account? Bond or cash should all move to BOXX. Did I miss something here? Thanks!
BND is a different asset class:
Longer term bonds with duration risk. And it’s all bonds, government and corporate.
But I agree: BOXX would be a very valuable substitute for BIL and SGOV.
For anyone interested. As of today’s 6mo T-bill auction the bps pick-up for a similar 6mo box spread in the OCT SPX was 24bps. Also, as regards all the comments on §1258, the experts I have spoken to have all opined that it is clearly a ‘conversion transaction’ BUT… it is a matter of visibility and desire to enforce. To date there hasn’t been much but the BOXX ETF issue may illuminate and motivate. Its similar to wash sale rules and moving between SPY/VOO/IVV/SPLG, If it looks, swims and quacks then……
Good point. It’s not on the IRS’s radar yet. Even if the S1256 advantage goes away, you can still make a small spread.
Fidelity’s rep says, “Fidelity no longer offers box spread strategies on our Multi-Leg Option ticket.” It appears they don’t allow cross-margining, rendering this unprofitable through Fidelity.
https://www.reddit.com/r/fidelityinvestments/comments/18do5gt/allow_crossmargining_for_long_box_spreads/
Thanks for the heads-up! Too sad to hear about Fidelity going that route! I use Fidelity for my basic brokerage needs (IRAs, Roths, etc.), but it appears that you need a more serious platform for box trades.
Thank you for all of the helpful information over the years! I have just started using your Google Sheet. I have a question about the Final Value Target field. If I put in 100, 50 or 25, all of the results stay the same? Shouldn’t the SWR percentages change based on the Final Value Target being different values? I am sure I am missing something or inputing the data incorrectly. Thank you very much for your help with this, Bernie
The SWRs should change if you change the final value target. You might have inadvertently changed/overwritten formulas.
Hi BigERN, Thank you for this post! Can you speak a bit more to the risk/potential loss of unwinding early? This seems to be another significant advantage over bonds or CDs, but I’m not sure I understand the magnitude of difference in risk there. If I needed to liquidate a 2-year treasury or CD early, that may be associated with a significant price drop or early w/d penalty (respectively). How does this risk compare with unwinding/partially unwinding a box spread? Is the box spread subject to interest rate risk, for example? Based on what you’ve said above, it seems like the risk/cost of unwinding early is somewhat nominal compared. Is that correct?
Unwinding early means you do the reverse transaction. Liquidity could be an issue. There might be wide bid/ask spreads and may not be able to unwind the position immediately. Also, if the market has moved and the strikes from the original box are no longer that liquid, it might be better to unwind with different strikes, see the section “How do I close the Box Spread?,” item 3.
Thank you. I have been waiting for an update on box spreads. And you knocked it out of the park. Nowhere else will you find such a thoughtful in-depth educational explanation.
Please accept my deepest gratitude.
Thanks! You rock!
The BOXX prospectus says: When purchasing or selling a Box Spread, the Fund will primarily use European-style options.
But then it says: The Fund expects to use options on the SPDR® S&P 500 ETF Trust for substantially all of the Fund’s holdings. Which from my understanding are American options.
If you go on the Alpha Architect website, at least 81% of the holdings are in SPY options. How big of a risk is this?
I checked here: https://etfsite.alphaarchitect.com/boxetf/#holdings
There are indeed some options on the SPY. But these are the European options. See the note below the table:
Thank you, good catch! Are these options special for institutions? I don’t see these tickers available on IBKR and TOS.
Yes, that’s what I suspect: I’ve not come across Euro-style options on ETFs on Interactive Brokers, so they may be special contracts just for the big guys (and gals).
Karsten,
Late question regarding box spreads. Do you find that it is better to match deltas on the call & put side (ie. I will buy the call (& sell corresponding put) and buy put (& sell corresponding call) at a 75 delta) OR match debit cost on each side (ie. same cost outlay on each of downside & upside) OR doesn’t it matter? Is the profit picture better with any of these options over the others?
The payoff diagram is the same, no matter what Deltas you pick, obviously. So this boils down to where you can get the best pricing. So, I would always put several box spreads on my watchlist tab at IB and see where the B/A looks best. Most of the time it’s pretty obvious. For example, for a 1000-spread, the 5000-6000 spread is the most active right now.
Thanks for the article Big Ern.
Is this a viable strategy with less than approximately $100k to lend?
You can do it with as little as $500, but the typical options + SPX exchange fees make it unattractive until you’re doing somewhere around $10k+ in my opinion.
Yeah, $10k might still work for long-dated spreads, so you can spread the roughly $6.50 transaction cost over a longer time period (e.g., 4-5 years)
My fees are only about $4.50 for each spread, but where to draw the line on what an acceptable % to lose to fees is subjective. I usually do 4 month spreads or longer.
If you get maybe 20-30bps extra yield p.a., then for a $10k principal $4.50 would eat up 4.5bps total or about 12bps over 4 months. That gets dangerously close to where it’s no worth the hassle.
You can do this Box Spread with much less. I don’t recommend it for less than $20k because transaction costs (about $6.50 to initiate) will eat up some of the gains.
Could ths strategy be used to borrow money instead of selling equities in a bear market? If you needed to it seems like you could do one spread a month dated a year out to cover that months expenses and roll them forward when they come due if the market is still down. or even one a year to cover the whole years worth of expenses. It would be like an interest only loan with low interest and borrowing a few years worth of expenses at 5% seems a lot better than selling off equities that are down 10%+. Am I missing something?
Please see this post about borrowing with the Box: https://earlyretirementnow.com/2021/12/09/low-cost-leverage-box-spread/
Yes, you can use the Box to borrow and it’s a perfect instrument: low rates and the implicit interest is tax-deductible.
And my SWR Series about borrowing against your portfolio instead of selling equities:
https://earlyretirementnow.com/2021/11/16/leverage-in-retirement-swr-series-part-49/
https://earlyretirementnow.com/2022/03/21/timing-leverage-in-retirement-swr-series-part-52/
The problem is that in a deep and long enough drawdown, you might get wiped out with a margin call. I would not recommend using only borrowing. Only partially and only when the market is down a lot!
Thank you! The information you provide and depth of knowledge and analysis is incredible. I have been on the path to FIRE for 10+ years and I have learned more in the 2 months since I discovered your site than in the prior 10 years.
Thanks for your kind words! That makes the work here worthwhile! 🙂