Disclaimer: This is not financial advice. Also, following these steps slightly wrong could result in you losing 100% of your savings. I have listed what I think are the most likely pitfalls, but still, beware.

Thanks to gilch and Mark Xu for helping edit this post, and Wei Dai for the original idea.

Summary

If you have a non-IRA brokerage account, you can borrow money on the open market at 0.55% and put it in a certificate of deposit (CD) yielding at least 1.25%. After taxes and commissions, this increases annual returns on your entire portfolio by 0.2-0.5%, nearly risk-free. This opportunity could continue for several years.

As this is $1000-2500 a year on a $500,000 portfolio, it's probably worth your time to set up if you're mid- to late-career, especially if you already do things like credit card churning. The administrative work includes making one trade every 1-3 years and a small amount of extra tax work.

Requirements

Before actually performing the trick, there are a few things you need:

  • A taxable (not IRA/401k) investment portfolio worth more than about $100k.
  • A brokerage account with portfolio margin and options approval. Portfolio margin is not strictly necessary but increases the return. I suggest using Interactive Brokers or TD Ameritrade.[1]
  • An account at an FDIC-insured bank (or NCUA-insured credit union) with a high-yield CD or savings account. You can find a list of such banks at depositaccounts.com, and approval is fairly quick. FDIC/NCUA insurance is limited to $250,000 per individual per bank, so if your portolio is over $500k you might need multiple banks.[2]

In total, the setup took me about 50 hours, but this involved lots of trial and error and double-checking; it can be done in 10 hours.

The Box Spread Trick

As of September 2020, some banks offer a 3-year CD yield of 1.25%, while the Treasury yield is only 0.2%. If you could borrow at the same rate as the US Treasury, you could make 1.05% per year risk-free. Sadly, you can't do this conventionally: mortgage rates are upwards of 2%, and most brokers charge upwards of 3% on margin loans. But you should be able to, because you have collateral in the form of a stock portfolio. This is where the box spread comes in.

A box spread is a combination of four options which cancel each other out so there is no risk from market movements.[3] This allows institutions to lend each other large sums of money: one party sells the box spread for a premium, and pays the loan back on the exercise date up to 3 years later. As a retail investor, you can't withdraw the cash from this loan, but you can take out a margin loan from your broker which is financed by this box spread. By using box spread financing, you pay market rates (typically treasury yields plus ~0.3%) rather than the broker's 3-10% margin rates.

To perform the trick, sell SPX box spreads in a total amount that's 30-65% of the value of your account depending on how frequently you want to monitor it. The exercise date should be as far in the future as possible, currently 2-3 years.[4] Each box spread is worth 100x the spread width; i.e. $10,000 for a 2900/3000 box. The legs should be 100-300 points apart to minimize commissions cost, and the upper strike price should be near-the-money (near the current value of the SPX) to maximize liquidity.[5] In current market conditions, the trade looks like this for a $250,000 portfolio; it's a loan with $150,000 due in December 2023.

SELL 5 DEC 15 2023 3000/3300 SPX BOX

Look up the current 2- or 3-year Treasury yield here, add 0.29%, and calculate the price, e.g. using this spreadsheet. Enter a limit order for that price.

WARNING: Don't trust a spreadsheet someone on the internet made. Double-check the math yourself. Also, make sure that:

  • You're using a LIMIT order (not market). Bid-ask spreads are absurdly wide.
  • You typed in the price correctly.
  • You're selling European-style options, e.g. SPX. This prevents early exercise risk.
  • You're selling the box, not buying it. (An order to sell a 3000/3300 box is equivalent to buying a 3300/3000 box; this is fine. The preview price should be negative, or say CREDIT).
  • You're doing a single trade, not four separate trades.
  • When previewing the order, the profit graph is a perfectly flat horizontal line, indicating that the options cancel out.

Execute the trade, and wait for it to be filled. If it isn't filled within a day or two, slowly walk up the price until it is. (I was filled at 0.32% above the treasury yield.) You may get a slightly better price or fill time by direct-routing to an exchange rather than using your broker's "smart routine". Once the order is filled, deposit the money into your bank account/CD. Your brokerage account should now have a cash balance near zero.

If the value of your investments declines enough during the 3-year period, you could be subject to a margin call and have your investments liquidated. Before this happens, withdraw some money from the CD (paying a small penalty, typically the last six months' interest), and redeposit it into your brokerage account.

Example

Let's say you have $350,000 in a Vanguard brokerage account, which is invested in mostly broad-based indices with some narrow-based indices, say mostly VT with some VWOB.[6] You get an Interactive Brokers account with options trading and portfolio margin, and pay a $50 fee to transfer your assets.

SELL 10 DEC 15 2023 3200/3400 SPX BOX

Say the Treasury yield on September 16, 2020 is 0.17%; you target a rate of 0.46%, and calculate the price. You make the trade above for $197,000 (0.471% APY); after a $26 commission, you get a $196,974 credit. The loan is due in 3 years for $200,000. You have already set up an account in a credit union with a 1.25% Jumbo 3 Year CD rate, so you withdraw the $196,974 from IB and use it to create a CD maturing in December 2023.

The liquidation value of your IB account is now $150,000 and your exposure is $350,000, a 43% ratio, comfortably above the 15% requirement. If the market declines by 28%, you could be subject to liquidation, so you set up a phone alert on the IB app.

In December 2023, the options are about to expire, so you buy them back or let them be exercised for $200,000. By then the CD has accumulated $8,164 in interest, which is $6,182 after-tax (your tax bracket is 24%) while the box spread has lost $3,026; this is $3,134 in profit. You then sell another box spread expiring in 2026, repeating the process.

FAQ

Is this really risk-free?

I'm moderately confident that the only major risks are (a) doing the setup wrong, and (b) having your options partially liquidated if the market plummets by 30%+ before you notice. Having part of a box spread liquidated is bad for two reasons: first, bid-ask spreads are wide, so you'll be liquidated at unfavorable prices, and second, holding part of a box spread is extremely risky, often equivalent to >20x leverage. You can limit this risk by limiting the size of the loan or using Interactive Brokers' "liquidate last" feature. You can eliminate it entirely by buying put options, though these are insurance and cut into your profit.

Why does such an arbitrage opportunity exist?

In short, people who put their money in savings accounts or CDs are less likely to withdraw it. Banks therefore give you higher rates. Wei Dai, who posted a comment outlining the trick in April, wrote that this effectively amounts to an FDIC subsidy. It's not quite free though, since you are giving up liquidity.

CD rates are currently 1% higher than treasury yields. This is higher than historical averages, but the gap previously stayed around 1% between 2009 and 2013. Because FDIC insurance on CDs is limited to $250,000 per bank per individual, a 1% gap probably can't be exploited at scale by institutions. The excess return could continue until market forces cause the gap to shrink, or until a large percentage of the money in high-yield CDs and savings accounts is from people using the box spread trick.

What if I pick stocks/own crypto/use Wealthfront/don't have portfolio margin?

The portfolio margin requirement for narrow indices and individual stocks is 15% rather than 10%. If you don't have portfolio margin, the Reg-T margin requirement is 25%. For cryptocurrency, no margin whatsoever is allowed. Actual margin requirements are often higher in volatile market conditions; sometimes brokers also have higher "house" margin requirements. Use your broker's tool to adjust the loan amount according to your margin requirement.

Passive services like Wealthfront don't support trading options, and so don't support the box spread trick.

Tax consequences?

I'm not a tax professional, but I think LEAPS (the long-term index options we're using) are Section 1256 contracts taxed as 60% long-term and 40% short-term capital losses. Interest from a savings account or CD is taxed as ordinary income, at up to 37%. The value of the box spread trick should decrease in higher tax brackets.

What if I'm outside the US?

The US has one of the strongest deposit insurance systems in the world. I haven't seen non-US banks with similarly strong insurance and high CD/savings yields.


  1. Interactive Brokers gives you options and PM approval on any account above $125k by checking some boxes. Rather than give a margin call, they immediately liquidate on a margin violation, but they offer a "liquidate last" feature so your options are not liquidated. They also offer lower margin rates, but this is not impactful because you don't use the margin loan. However, their software is outdated and less user-friendly. TD Ameritrade has better software and (as of 2020) also has a bonus for transferring your assets there, but requires that you pass a written test on options. Other brokers might work (leave a comment if they do), but I've heard that Robinhood doesn't let you trade box spreads. ↩︎

  2. NCUA and FDIC are both backed by the full faith and credit of the US government, so the risk of bank failures is small. If you have a spouse, you can get one account for each of you, plus a $500,000 joint account; this lets you have $1M of total FDIC/NCUA insurance at just one bank. ↩︎

  3. The box spread is explained further at the links in this comment. ↩︎

  4. December options start trading in the September 3 years prior. ↩︎

  5. I have heard that institutions use a 1000/2000 box, but these can only be exchanged in increments of $100,000 and so could be a bit unwieldy. ↩︎

  6. This is a reasonable asset allocation if you mostly believe in EMH and want to approximate the global market portfolio but already own real estate and think Treasury yields are unreasonably low. ↩︎

New Comment
44 comments, sorted by Click to highlight new comments since:

To check I understand the mechanics of this, let me try to restate parts of it:

  • A box spread is a combination of options engineered to be worth a fixed amount at a specified future time. Thus, buying a box spread works out like lending money, and selling a box spread works out like borrowing it.
  • I think the specific choice of SPX (the S&P 500) here doesn't matter too much, but presumably it'll be good because it's something that gets traded on a lot.
  • If you borrow money like this, you can't in general withdraw it, because your broker doesn't know you'll be able to pay it back. But if you already have a lot of money invested with your broker, they might let you do so, using those investments as collatoral. The caveat will be that if your current investments drop below a certain value, your broker will ask you to pony up some more collatoral in the form of cash, possibly selling some of your investments.
  • Using this trick, you sell a box spread to borrow money at a lower interest rate than you can normally. Then you withdraw it and invest in a different account, which has a fixed interest rate higher than the rate you're paying to borrow the money.

Does that sound about right? The third point is the one I'm least sure about.

(I'm unlikely to make use of this myself, not living in the US and not having that kind of money to invest. I just want to understand this stuff better than I currently do.)

I think the specific choice of SPX (the S&P 500) here doesn't matter too much, but presumably it'll be good because it's something that gets traded on a lot.

The liquidity isn't the only reason. SPX has European-style options. If you tried this on SPY instead, you'd be exposed to early-assignment risk. SPX is not the only index with European options, but it is the most liquid.

I'm unlikely to make use of this myself, not living in the US and not having that kind of money to invest.

You can adjust the width of the boxes to borrow a smaller amount. And the XSP mini options are one-tenth the size of the SPX options. They're not as liquid though. Some brokers take international customers.

To explain/sanity check why I think it doesn't make sense for me personally to do this:

A thing the UK has is ISAs, which are accounts that you can put a certain amount into per year and not pay tax on your gains (or interest, for fixed-interest ISAs). The limit is something like £15k. One of my ISAs lets me invest in various stock indices, but not individual stocks, and I'd be surprised if there was one that let you pull off a trick like this.

Separately, any money I put into my pension lets me reclaim the income tax I already paid on it. Then I think I need to pay tax when withdrawing it, but some of it is tax free and the rest will be at a lower rate unless I withdraw a lot or have other income. (Or I guess if the rules change, which is maybe a possibility I should take more seriously.) My pension also lets me invest in stock indices.

So as I understand it, the bulk of what I invest should (and currently does) go to an ISA or pension. It's probably not harmful to put moderate amounts in the kind of broker that would let me do this trick, because I only pay capital gains tax if my gains are above some threshold. But I suspect the intersection of "enough money it's worth doing this trick with" and "not so much that I should just have it in an ISA" is either empty, or narrow enough to be only barely worthwhile.

(I'd also need to figure out where to put the money I've borrowed. We have current accounts that pay a few % on up to a few £k, but for large amounts the best safe investment I can think of is maybe premium bonds. That's a lottery that pays 1.4% interest on average, tax free. You can invest up to £50k in them. But then doing it with a smaller amount is riskier, because there's more chance that I don't earn enough.)

That's more or less right and clearer than how I wrote it up. Two slight nuances to the third point which I don't know if you understand correctly:

(a) the broker allows you to withdraw all but 10-50% of the value of your investments in a margin loan even without a box spread; they just charge exorbitant interest rates since they're financing the loan themselves.

(b) a box spread has two credit legs and two debit legs; if the value of your investments drops, the broker might sell the debit legs of the box spread rather than your other investments, which exposes you to extreme levels of risk.

Thanks, I didn't know these.

the broker allows you to withdraw all but 10-50% of the value of your investments in a margin loan even without a box spread; they just charge exorbitant interest rates since they’re financing the loan themselves.

To clarify, the difference between this and the box spread loan is:

  • With the box spread, your brokerage account gets deposited say $100k cash that someone else has lent you. The brokerage knows you'll need lots of cash in future, so they need collatoral to let you withdraw it. When you do, the value of your account will be $100k lower than it was before.
  • With this, your brokerage account has say $100k invested, but it's not currently in the form of cash. Your brokerage is willing to lend you $10-50k cash, and let you withdraw it; but they need both collatoral and interest for that. When you withdraw it, your account will still be worth $100k.

?

Not sure I've understood your question, but if you withdraw from your account, the value goes down, whether you've sold a box spread or just have a margin loan.

Hm. "Value" might not have been quite what I meant there. The thing I was pointing at was that if you start with $100k worth of google stock in your account, you'd (if I'm understanding this correctly) still have $100k worth of google stock, even if the brokerage also marks you as having a $10-50k liability. You might have to liquidate some of it in future if there's a margin call, but your goal is presumably to avoid that.

In light of this, to clarify the first point - "When you do, the value of your account will be $100k lower than it was before" - I meant $100k lower than it was before withdrawing, because previously your account had $100k cash and now it has $0k cash. I was thinking it would be the same value as it was before you sold the box spreads. But if selling the box spreads doesn't change the account value much - if the brokerage marks you as having $100k cash but also a ~$100k liability - then after withdrawing, the value is lower than before you sold the spreads.

But if selling the box spreads doesn't change the account value much - if the brokerage marks you as having $100k cash but also a ~$100k liability - then after withdrawing, the value is lower than before you sold the spreads.

This is correct. The brokerage shows cash + a liability.

This trick gives an interesting way to borrow money for cheap, and I could imagine other uses for it than CD deposits.

The article gives the impression that the only use one can make of the box spread trick is to invest the proceeds in a CD. But one could as well invest them in crypto or to fix their house, etc. Now regardless of the use the borrower chooses, it's their responsibility to make sure they can still repay it soon in case of margin call. I can think of at least 3 ways other than CDs that accomplish this.

This is a really valuable article! Thank you for writing it. 

For me, trying to execute on this strategy has hit and miss. Using Interactive Brokers, I was able to execute a Dec14'23 3500/4000 box spread on Jan 12th 2021 (1066 days before expiration) for a cost of $488.50, generating an effective interest rate of 0.81%. This isn't a horrible interest rate, but the 3 year treasury yield on Jan 12th was 0.23%. Adding 0.3% to the treasury yield, I would have expected this transaction to go through at 0.53%, not 0.81%. Since then, I've tried different types of box spreads (3000/4000 box, 3000/5000 box, 1000/2000 box), and I can't get an execution at (three year treasury yield + 0.4%) or lower.

Questions for the group:

  1. This article says executions should occur at 0.29% above treasury yields. Is anyone able to consistently get executions at that rate?
  2. In Interactive Brokers' TWS software, only "Smart" routing is available to me for building these 4 option legs. Customer service confirmed that only smart routing is available. I can't select manual routing. Can someone confirm that they can manually route 4 legged options using Interactive Brokers, and if so, how?
  3. What is the best broker for doing these kinds of transactions?

I've been able to get closer to 0.6% on IB. I've done that by entering the order at a favorable price and then manually adjusting it by a small amount once a day until it gets filled. There's probably a better way to do it, but that's what's worked for me.

What box spread upper and lower bounds have you been using? Are you using IB's smart routing or manual routing?

I used SPX Dec '22, 2700/3000 (S&P was closer to those prices when I entered the position). And smart routing I think. Whatever the default is. I didn't manually choose an exchange.

This article says executions should occur at 0.29% above treasury yields. Is anyone able to consistently get executions at that rate?

A few weeks ago I was filled at about 0.51% above treasury yield, which is definitely worse than the 0.29-0.35% I previously got. Both were using IBKR smart routing and there weren't any obvious differences, so I'm guessing box spread rates are driven by supply and demand, and there's just relatively more supply now than in mid-2020.

After having an order entered for about 3 days, I got an execution today (Jan 22 2021) on a SPX 1000/2000 15DEC23 box for 0.65%, which is 0.46% above the 3 year treasury yield. This was using IB's smart routing (manual routing not available). 

Does anyone have thoughts on whether buying Treasury Inflation-Protected Securities (probably in the form of an ETF) on margin would be a good way to hedge against currency devaluation?

Buying RINF (an etf) would likely be similar but better. There are plenty of other less direct ways, such as gold or cryptocurrencies.

In August 2019 I set up a short box spread on the EOE index (AEX). I used options with 200 and 800 strike prices with strike date in December 2023. Setting up the box yielded EUR 61230,- per box and in 2023 the box will settle at EUR -60000,-. This yield was possible due to the negative interest rate so this was an interesting way to borrow money with negative interest. This way I built a virtual deposito with positive yield for my cash. 

This box is intrinsically risk free up to a certain level of market stress. When Covid hit the markets in March 2020 the volatility rised skyhigh resulting in crazy spreads which made trading impossible and which caused my broker to change their risk management. This resulted in devaluation of my stock portfolio which I use for coverage. I admit that the way this broker calculated risk was slightly irrational but it's their call. I covered the margin calls by depositing extra cash into my account. 

Another risk I encountered is that interest rates may change. In 2020 my broker announced that they would apply negative interest rates for cash positions larger than EUR 100000,-. This may consume the yield which would tempt you to invest cash which may not always be the best choice. 

There are upsides too: Today, (Feb 2022) the interest rate (EUR) is rising (NLGGB10J.BONDS) which results in an increasing result for this box as it now yields less than EUR 60000,-. As it looks now this box will develop succesfully, I expect to liquidate this box before strike date. A truly free lunch. 

Nice information Hubertus, would you still recommend a box spread in today's market, and would one be possible with a 4 year CD, I am located in the USA and the 3 year CDs are all at 1.25% however a 4 year CD from NASA FCU is at 1.7% as of Feb 1, 2022. 

I don't recommend borrowing money to invest it. 

In today's market I would not set up the virtual deposito I mentioned above, it would not work. 

At this moment (Feb 8 2022) the short box spread AEX 200/800 DEC 2023 yields EUR 59944,-, with settlement in 2023 at EUR 60000,- this would result in a loan at 0,09% interest.  This is nowhere near the negative % I got in 2019. However, with investing in a CD at 1,7% you may think that gives a nice leverage.

By investing the yield in a CD you won't have the broker's cash interest risk. But as long as a broker decides about risk management, stock portfolio valuation and margin rates I would not consider this a safe loan for long term investments. There is a serious risk that should (can?) be covered. Interesting stuff. 

Again, I don't recommend borrowing money to invest it, actually, I don't recommend investing at all  ;-)

Thank you so much for your explanation. I invest very conservatively so have never leveraged or invested with borrowed money, but have been looking to invest a little more aggressively as I am young and would only do so with a small portion of my portfolio. I have never traded options so this is very new to me and interesting to learn and read about.  Thanks again!

Just curious, what maintenance margin do IBKR/Ameritrade require for this box spread? 

I suppose they might depend on the nature of the collateral, so let's assume my collateral is 100% SPX.

i.e. if I were to max out the box spread loan amount, what's the maximum percentage that my collateral could drop before IBKR issues the margin call?

Depends whether you're holding it as SPY/VOO or as a mutual fund (assuming you don't mean SPX futures, and SPX itself isn't tradeable). IBKR has somewhat higher margin requirements than the minimums set by the OCC; I think they require 25% maintenance margin for mutual funds and 10-15% for broad-based stock indices.

Worth noting that you may also want to use this trick if you are using portfolio margin to lever up your portfolio more than 1x. Instead of withdrawing to a CD, you leave the cash in your account, and end up with a lower margin rate than you'd otherwise get.

One important warning to add about box spreads: AVOID options that expire on the last two business days of the year. They can result in a huge tax burden and/or make filing taxes a huge pain.

Why? Because long options are taxed in the year they close, while short options are taxed in the year they settle. With settlement taking two business days, you can end up with the long and short legs being taxed in different years. If the longs have a gain and the shorts have a loss, you're stuck paying taxes on the gain and waiting another year (or more in some cases) to claim the loss.

"But these are Section 1256 contracts," you say. "Any unrealized losses will be marked to market and applied to the current year." Yes, they should be. But your broker (*cough*, TD Ameritrade, *cough*) might have other ideas. After hours of back and forth, they'll get stuck on the following:

"We can't give you realized P&L because the trade settled in January."
"And we can't give you unrealized P&L because the trade was already closed at end of year."

You'll be stuck either accepting the tax burden or figuring out how to work around an incorrect 1099 (which the IRS will have received from your broker).

Note that these can be very large amounts -- the taxable gain can be much larger than the credit you received for the box spread.

Here's a Reddit post of someone else who also ran into this issue: https://reddit.com/r/options/comments/ddokzu/taxation_for_options_contracts_closed_on_last/

(Would you consider updating the post to mention this?)

The linked WSB post alludes to one more obscure risk with IBKR in particular,

Portfolio margin traders can suffer margin calls if you have a bad mark/quote. This is a huge problem at Interactive Brokers...You need to place a GTC close order at a very favorable unrealistic price to always quote your box.

Brief Googling does indicate there's some history of IBKR customer disputes about liquidations triggered by some weird, transient price tick in a thinly-traded contract. I'd like to think that wouldn't happen in SPX of all symbols, but, we ought to imagine sh*t hitting the fan in period of market duress...

I don't fully understand the suggestion in the last sentence. Can somebody explain how keeping an open order to close the combo position is supposed to help? Other ideas?

I'm confused too. Say you sold 10 3200/3400 box spread contracts for $196k a while ago, for a "par value" of $20k per box or $200k total. Since the exact box spread you sold is rarely traded, the best bid in the order book will be the sum of the 4 individual leg bids, and likewise the best ask will be the sum of the 4 leg asks. This makes the spread very wide, maybe $15k bid and $25k ask, and I'm not sure how IBKR values your position, maybe it just takes the midpoint. Let's say it's normally -$197k at some point in time.

Now say someone buys a 3200/3400 box for $25k. I'm guessing that IBKR's algorithm could see the last traded price, decide your position is worth $-250k now, and liquidate you. If you keep an open order to buy back the box for say $16k each, nothing changes and in fact the midpoint price would increase. Maybe the commenter meant to say you should keep an open order to sell more box spreads for slightly above par value?

Here's a thread I found with some interesting information (albeit by anonymous Internet commenters :) It ends up, on page 5, with a suggestion similar to yours -- open a GTC order to add to (not close) your position, at a price that won't execute, but isn't so crazy as to imply a margin violation were your position to be marked there.

Now, I don't see any harm in putting out there an offer to borrow at negative interest; but it's still not obvious to me why this truly addresses the risk. Do we need such orders to backstop each individual leg as well as the combo as a whole? Since the IBKR auto-liquidation algorithm is proprietary and evolving over time, there seems to be no way to know for sure.

--

One more important info for other IBKR users: Set Liquidate Last is no longer available for legs of a combo in the main Monitor view (it's grayed out). Support replied to me: On further review the ability to set an individual leg on a combo position to "Set Liquidate Last" is no longer available. I apologize for any confusion this has caused via the information on our website.

Actually though, I found that you can still Set Liquidate Last the long legs by going through the Account Window - Portfolio list instead of the Monitor view. Judging from support's reply, their programmers probably just forgot to disable the menu item there too, so I wouldn't believe it.

--

After investigating these tail risks, I personally don't find that CD interest arbitrage provides enough reward for having to carry these small worries in the back of my mind for several years. But the box spread financing is definitely a cool idea to have holstered, just in case a unique opportunity (with more upside) comes along for which I'd need a medium-term bridge loan...like MMM's impulse purchase, which led me to this article in the first place. Thanks!

Isn't another risk if the market tanks within the first few months, because you will have to pay the withdrawal penalty from the CD out of pocket before you have the interest accumulate? This risk seems proportionate to the benefit (given that we have more than one huge correction every 50 years, there is a > 2% chance that the market will have a huge correction in the first year).

You say that you are moderately confident that the risk did not include this case, so I'm likely missing something (or you need to do a moderate update).

Another thing I'm still curious about is who the buyers of these box spreads are (assuming the legs are sold as a combination and not separately) . The discussion says the arbitrage opportunity comes from the FDIC but the FDIC does not directly buy the spread; they allow the CD to exist, which the buyers should have access to. So do the buyers have access to options but not CDs, or are there other advantages that they have which I am missing?

the FDIC does not directly buy the spread; they allow the CD to exist, which the buyers should have access to

I'm not sure how important FDIC insurance is to the story here, but worth noting that it has a 250k per account limit. So I don't think financial institutions would have access to an unlimited amount of it.

The risk is small, because for most CDs the withdrawal penalty is small. My credit union allows partial withdrawals and charges a fee of 6 months' interest (about 0.625%) on the amount withdrawn, so if the market tanks 30% and you have to redeposit say 20% into the margin account, you have lost a negligible 0.125% and the box spread trick still comes out ahead for the year. 6 months' interest is typical. It's important to choose a bank or credit union that has reasonably lenient terms, though.

The risk is small, but so is the benefit. As a result this was not a trivial analysis for me. I doubt if it's low risk to redeposit just 20% at a 30% market drop due to volatility (the recent intraday crash movements exceeded 10%, and you get a margin call at a 40% with the 250k/150k example). After mulling it over I think I agree that it is worth it anyways though.

Here's some more examples that I ran over. A better person would figure out the probabilities and equations to integrate over to calculate the expected value.

Withdrawing 75% at 0 months should be the break even (you lose .625% * 37.5k = 700 to penalties, but you get back a bit more than that with interest of .625% * 112.5k * 3 years = 700).

If you don't need to withdraw within the first 6 months, you are ahead. Let's look at the 100 percentage withdrawal case first to keep it simple. So the actual break even is withdraw 100% at month 6, and any time before this you lose money. If this happens at day 0 you lose .625% * 150k, or .375% of your portfolio as the worst case scenario. So the interesting loss scenarios range from 75% immediately to 100% in 6 months.

I don't have the probabilities of these events happening, but casually looking at it, it seems like they happen at a significantly lower probability than the equal or greater gains case. Though it should reduce your expected value by a small amount.

I'm not sure if I completely understand the stock drawdown calculation part - so I tried to walkthrough your example in reverse. 

Starting with the $350,000 portfolio, you borrow $200,000 and reinvest that amount into the same equity. So when the equity declines by ~18%, you are left with $451,000. Since you still owe $200,000 my understanding is that the maintenance margin calculation would be as follows:

$251,000 / $451,000 = 56% > 15% 

Therefore, no margin call occurs.

Since I'm still new to this I'm sure I have a logical blunder embedded somewhere. Could you correct my misunderstanding?

Your scenario seems totally different from the one described in the post. You're keeping the $200K loan in your brokerage account (not withdrawing it) and plowing it back into the risk asset (not a risk-free CD); then imagining only an 18% decline (not 28%).

OP's liquidation scenario occurs with a $200K liability against $252K equity in the brokerage account (following the 28% decline from the original $350K). The brokerage let OP withdraw the $200K, but then can't assume he still has it.

Fixed annuities are a good alternative to CDs. Better yield, worse creditworthiness. 3 year MYGAs near 2.4% 8/17/2021.

Is it unlimited? If I have $100 in my IB account, can I sell $1,000,000 box?

Why would you? The point of the box spread trick is to reduce interest on margin loans, they don't increase your buying power. Selling box spreads without a margin loan just makes you pay small amounts of interest for no reason.

How much buying power does IB require if I want to sell $10,000 box?

Are there any details to be concerned about how this settles at expiration?

For example, is the following all correct? However the market moved,

  1. one of the short legs is assigned for you to settle with cash -- possibly an enormous amount thereof!
  2. one of the long legs is in-the-money by the same amount, minus the loan value; and exercises automatically (assuming you didn't specify contrary instructions)
  3. the other two legs expire worthless
  4. your only required action is to ensure the loan value is on-hand in the account (or within margin borrowing power) at this time

Could anything possibly go wrong here?

I haven't let one expire yet, but that looks right. I don't anticipate any major problems because none of the sources I've seen mention any, and besides all legs expire on the same day.

Note that the loan value already has to be within margin borrowing power throughout the entire duration of the loan, since your broker takes into account the box spread as a negative-value asset when calculating margin. Box spreads can't get you more borrowing power (it's box spread financing, not box spread borrowing).

With tax season coming up, I am curious to see how your short box spreads are being treated on your taxes. 

In the case of your 3200/3400 box example, my understanding is that these are 1256 contracts which as of Dec 31 resulted in a mark-to-market gain of 200K to your account treated at a 60 long/40 short rate. That's a whole lot of imputed income for a set of unrealized gains! 

My understanding is that each leg of the box spread is marked-to-market, resulting in a small overall capital loss equal to the interest you're paying to the holder of the long box. For example, the market is up this year, so the bullish legs gained value and the bearish legs lost slightly more value. This is how the Interactive Brokers tax forms work; on the section-1256 section of the 1099 form, the line "Unrealized profit or (loss) on open contracts - 12/31/2020" has a loss of about $2000. If interest rates rise so much during a year that the interest is negative, I would expect there to be a small 60/40 capital gain. I could be missing something though.

Why is the financing rate measurably above the risk-free rate? My understanding is that the box spread replaces bank or broker loan or deposit interest (who make a profit from the spread between interest paid and interest received). The box spread is arm's length, i.e. cuts out the middle man. Retail investors and institutions both use the same options market. Shouldn't the implied rates of any options combination be close to the risk-free rate? If I buy the box instead of selling it, can I make ca. 0.5% above the risk-free rate?