Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
airbear
Posts: 1
Joined: Fri Nov 12, 2021 6:04 pm

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by airbear »

First post - been lurking on these forums for a bit now.

How much capital should one have before considering jumping from HFEA (or HFEA with TYA) to mHFEA using futures?
DMoogle
Posts: 549
Joined: Sat Oct 31, 2020 10:24 pm

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by DMoogle »

The net value of the portfolio I'm doing this with is $480k. Current allocation is around $700k in equities and around $1.2M in ZN (9 contracts). I have a very high risk tolerance. ZN got fairly wrecked the past quarter (my portion on it is down $27k since I started on it), but whatevs - I agree with the fundamental concepts behind this strategy, and it's not for the faint of heart.

So when I roll over my futures in the coming month, current thought is to (based on the contract prices above) liquidate the ZN contracts for maybe:
5 contracts ZF ($484k)
3 contracts ZN ($390k)
1 contract TN ($145k)
1 contract ZB ($161k)
For a total of $1.18M exposure (similar to now), and probably keep my current equities (or tax loss harvest if I can).
Investing Lawyer
Posts: 51
Joined: Wed Apr 08, 2020 9:35 pm

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by Investing Lawyer »

DMoogle wrote: Fri Nov 12, 2021 6:49 pm The net value of the portfolio I'm doing this with is $480k. Current allocation is around $700k in equities and around $1.2M in ZN (9 contracts). I have a very high risk tolerance. ZN got fairly wrecked the past quarter (my portion on it is down $27k since I started on it), but whatevs - I agree with the fundamental concepts behind this strategy, and it's not for the faint of heart.

So when I roll over my futures in the coming month, current thought is to (based on the contract prices above) liquidate the ZN contracts for maybe:
5 contracts ZF ($484k)
3 contracts ZN ($390k)
1 contract TN ($145k)
1 contract ZB ($161k)
For a total of $1.18M exposure (similar to now), and probably keep my current equities (or tax loss harvest if I can).
How are you calculating how many of the other contracts to hold? Are you using the ratio of duration from/to ZN? Like you, I currently hold all my exposure in many ZN contracts.
DMoogle
Posts: 549
Joined: Sat Oct 31, 2020 10:24 pm

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by DMoogle »

Investing Lawyer wrote: Fri Nov 12, 2021 9:57 pmHow are you calculating how many of the other contracts to hold? Are you using the ratio of duration from/to ZN? Like you, I currently hold all my exposure in many ZN contracts.
Swagging it. comeinvest posted that he has, in order of preference, ZF > ZN > TN > ZB. This feels about right - majority of the concentration is ZF/ZN, which focused on ITTs, which has largely been shown to be optimal (with maybe some debate in regards to STTs). Adding some duration diversification with TN/ZB can't hurt that much.

Nobody knows the perfect answer here. I prefer to think of it as being directionally correct, with room for optimization improvement. Since the futures contracts need to be rolled over every quarter anyway, I think that's a good way to double-check the current best thinking and make adjustments as needed.
Bread Investor
Posts: 7
Joined: Sun Nov 07, 2021 10:44 pm

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by Bread Investor »

Thanks Klaus14 and skierincolorado for your inputs. With your inputs, as well as reading through most of the posts in this thread, I've decided on a 165/250 Stock/ITT allocation.

Would the following work for a $100k portfolio?

Image

Some additional questions (I'm using Interactive Brokers) which I need your advice:
1. How often should I do the rebalancing (e.g. Monthly/Quarterly)?
2. For futures (i.e. MES & ZF), is the rule is to always buy the contract that is nearest to expiry, and then roll over around a week before expiry date?
3. Is there a way to automatically roll over the futures contract, instead of having to manually do it quarterly? I see that Interactive Brokers has an "Auto-roll feature in this link: https://www.interactivebrokers.com/en/s ... llover.htm", but it seems to be only for futures data? Does it auto-roll for my futures position too?
Hfearless
Posts: 135
Joined: Sat Aug 14, 2021 9:53 am

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by Hfearless »

airbear wrote: Fri Nov 12, 2021 6:09 pm First post - been lurking on these forums for a bit now.

How much capital should one have before considering jumping from HFEA (or HFEA with TYA) to mHFEA using futures?
If the assumptions underlying mHFEA are correct (which is the biggest black swan generator here), mHFEA is superior to HFEA regardless of total amount. However, it requires more leverage than HFEA and LETFs can only provide so much of it, hence the futures. The obvious problem with the futures is that they come in $100k+ chunks.

If you have much capital but are only willing to dedicate a small part of it to mHFEA, instead consider buying futures for all your capital in the correct ratio (what exactly that is is subject to debate) and leveraging your entire portfolio but not to a high level (e. g. at a leverage level similar to that of NTSX).
perfectuncertainty
Posts: 386
Joined: Sun Feb 04, 2018 6:44 pm

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by perfectuncertainty »

comeinvest wrote: Wed Nov 10, 2021 4:32 pm P.S.: Skier created this thread, which is my favorite on the BH forum! He also selflessly keeps explaining his valuable findings to fellow investors. It is not his fault that ITTs dropped as of late.
I think Skier is 100% responsible for the ITT performance :D :D :D :D :D

Seriously, Skier has caused me to think about HFEA differently.
DMoogle
Posts: 549
Joined: Sat Oct 31, 2020 10:24 pm

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by DMoogle »

Bread Investor wrote: Sat Nov 13, 2021 12:43 amSome additional questions (I'm using Interactive Brokers) which I need your advice:
1. How often should I do the rebalancing (e.g. Monthly/Quarterly)?
2. For futures (i.e. MES & ZF), is the rule is to always buy the contract that is nearest to expiry, and then roll over around a week before expiry date?
3. Is there a way to automatically roll over the futures contract, instead of having to manually do it quarterly? I see that Interactive Brokers has an "Auto-roll feature in this link: https://www.interactivebrokers.com/en/s ... llover.htm", but it seems to be only for futures data? Does it auto-roll for my futures position too?
I'm still figuring out some of the logistics myself, so take these answers with a small grain of salt:
  • In HFEA, rebalancing quarterly had the best performance. You need to roll over futures once/quarter anyway, so may as well do your rebalancing then. From a fundamental/conceptual POV, I don't think it matters that much.
  • You don't have to, but that's when volume is highest, so spreads should be lowest. Seems like the best way to go.
  • I was very confused by that as well, but as far as I can tell, that "auto-roll" is only for data purposes, not for execution. So yes, you need to roll manually (sell current contracts, buy contracts for next quarter).
One other point: did you calculate how much collateral you need for those futures positions? At a quick glance, $67k seems like way more than you need, so I would swap some of that MES for a domestic equity ETF of your choice. Reason being: MES has an implied financing cost (something like 0.4% IIRC), whereas the ETF will only have its expense ratio.
Topic Author
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by skierincolorado »

DMoogle wrote: Fri Nov 12, 2021 6:49 pm The net value of the portfolio I'm doing this with is $480k. Current allocation is around $700k in equities and around $1.2M in ZN (9 contracts). I have a very high risk tolerance. ZN got fairly wrecked the past quarter (my portion on it is down $27k since I started on it), but whatevs - I agree with the fundamental concepts behind this strategy, and it's not for the faint of heart.

So when I roll over my futures in the coming month, current thought is to (based on the contract prices above) liquidate the ZN contracts for maybe:
5 contracts ZF ($484k)
3 contracts ZN ($390k)
1 contract TN ($145k)
1 contract ZB ($161k)
For a total of $1.18M exposure (similar to now), and probably keep my current equities (or tax loss harvest if I can).
I think this is equal risk in ZB as ZF. I think the weighted duration would be just over 8 years.

Even during periods of flattening when the 8 year rate dropped more than the 5 year rate, the 5 year rate has significantly outperformed. I'd try to keep the weighted duration somewhat close to 5. During periods of non-flattening, the outperformance of 5 year has been even greater. I think it's fine to go a little longer than the sweet spot historically if that is personal preferance, but 8 years is well outside of that zone. DIversifying makes sense, but the weighted duration should stay below 6 or 7 which is still on the longer side but shouldn't make too much difference. Could things change in the future? Sure - but we'd have to see more of a term premium and slope to the curve beyond 7 years *before* that can happen. Not only does history tell us that a weighted duration of around 5 is better, but it's also very likely to be true based on the current shape of the yield curve. The only scenario where the longer durations win is if the curve flattens rapidly, which would be the demise of this entire strategy anyways since no profit can be made leveraging a flat curve.

Since dec 31 2002, the 10 year interest rate dropped 0.7% more than the 5 year rate, making this a favorable period for IEF over VFITX, and yet VFITX still wins solidly.
https://www.portfoliovisualizer.com/bac ... on3_2=-120

A more neutral period would be Oct 2004-present, in which the VFITX win is even more solid. I'm not saying we all have to be exactly like VFITX, somewhat shorter or longer durations could be reasonable. More diversification could be beneficial. I just personally wouldn't stray too far. It's also a bit of a risk to change the AA so drastically. You'd be increasing your current duration by nearly 50% - a big switch in your positioning on the curve. And having already suffered through a period of rapid flattening, it might not be an opportune time to make the switch (especially given history argues for a duration shorter than ZN not longer). It would be unfortunate to bite the bullet on the last 2 months of flattening, only to switch into longer durations and see the curve steepen again. My advice would be to stay the course on your current duration, but diversify if desired.

https://www.portfoliovisualizer.com/bac ... on3_2=-120
Last edited by skierincolorado on Sat Nov 13, 2021 10:42 am, edited 2 times in total.
Topic Author
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by skierincolorado »

Bread Investor wrote: Sat Nov 13, 2021 12:43 am Thanks Klaus14 and skierincolorado for your inputs. With your inputs, as well as reading through most of the posts in this thread, I've decided on a 165/250 Stock/ITT allocation.

Would the following work for a $100k portfolio?

Image

Some additional questions (I'm using Interactive Brokers) which I need your advice:
1. How often should I do the rebalancing (e.g. Monthly/Quarterly)?
2. For futures (i.e. MES & ZF), is the rule is to always buy the contract that is nearest to expiry, and then roll over around a week before expiry date?
3. Is there a way to automatically roll over the futures contract, instead of having to manually do it quarterly? I see that Interactive Brokers has an "Auto-roll feature in this link: https://www.interactivebrokers.com/en/s ... llover.htm", but it seems to be only for futures data? Does it auto-roll for my futures position too?
A few suggestions: a) do 1 ZF and 1 ZN. The duration of ZF is more like 4.5 years, and the backtests I've posted mostly use ETFs with 5-5.5 year durations. Plus you'll get a little diversification from holding one of each. b) You are near the cutoff for dropping an MES and buying 23k of VTI. You would still hold 44k of cash collateral for 5 MES + 1 ZF + 1 ZN, which is near 7k per contract. c) If you decide to do (b) you could buy VB and/or VBR to make up for the large cap tilt of MES.

Regarding your questions:
1. There has not been much discussion of rebalancing. I would assume similar to HFEA would work best. So far I have just used new funds to rebalance.
2. Yes - although for the treasuries IBKR forces you to roll just over a month before expiry
3. I don't think so. Confusingly that auto-roll is for the data not for the actual investment.
Hydromod
Posts: 1048
Joined: Tue Mar 26, 2019 10:21 pm

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by Hydromod »

skierincolorado wrote: Sat Nov 13, 2021 10:33 am https://www.portfoliovisualizer.com/bac ... on3_2=-120

https://www.portfoliovisualizer.com/bac ... on3_2=-120
Broken links...
DMoogle
Posts: 549
Joined: Sat Oct 31, 2020 10:24 pm

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by DMoogle »

Thanks for the analysis above, skier. I'll retweak based on your suggestions.
skierincolorado wrote: Sat Nov 13, 2021 10:39 am2. Yes - although for the treasuries IBKR forces you to roll just over a month before expiry
Interesting, I didn't know that. I guess the Dec ones will be right around the corner then?
klaus14
Posts: 953
Joined: Sun Nov 25, 2018 6:43 pm

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by klaus14 »

DMoogle wrote: Sat Nov 13, 2021 2:55 pm Thanks for the analysis above, skier. I'll retweak based on your suggestions.
skierincolorado wrote: Sat Nov 13, 2021 10:39 am2. Yes - although for the treasuries IBKR forces you to roll just over a month before expiry
Interesting, I didn't know that. I guess the Dec ones will be right around the corner then?
It is also the most optimal to roll them in the last days of the prior month. Everyone rolls around that time and you have very good liquidity. My notes say: IBKR liquidates on second last business day of the prior month (Nov 29) but the most optimal is -4 (Nov 25) or -5th (Nov 24).
My investment algorithm: https://www.bogleheads.org/forum/viewtopic.php?f=10&t=351899&p=6112869#p6112869
keith6014
Posts: 697
Joined: Thu Jan 02, 2014 10:58 pm

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by keith6014 »

adamhg wrote: Sun Nov 07, 2021 9:44 pm Thinking about this more, options might be a good way for short term treasury exposure. IV is so low that theta is effectively zero.

Take the Jan 2023 $80 call. Theta is around 0.00005, $5.9 midpoint and a 0.99 delta. That would give us a leverage ratio of about 13x (.99×80÷5.9) and an "implied financing cost" of 0.0031% (.00005×365÷5.9) and never having to deal with margin maintenance.

Anoher benefit of options is that for tail risk, when the market tanks IV would spike which would theoretically make the optima more valuable than simply the underlying. Of course the opposite is true too.

I might actually try this with a small account.
Just writing to say thankyou to SIC and AdamHg. Please keep us posted on how it goes with options.
Lock
Posts: 98
Joined: Sat Oct 05, 2019 8:52 pm

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by Lock »

https://www.cmegroup.com/trading/intere ... oftheroll/#

This is the best resource for when to roll your treasury futures.
Bread Investor
Posts: 7
Joined: Sun Nov 07, 2021 10:44 pm

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by Bread Investor »

DMoogle wrote: Sat Nov 13, 2021 10:19 am
Bread Investor wrote: Sat Nov 13, 2021 12:43 amSome additional questions (I'm using Interactive Brokers) which I need your advice:
1. How often should I do the rebalancing (e.g. Monthly/Quarterly)?
2. For futures (i.e. MES & ZF), is the rule is to always buy the contract that is nearest to expiry, and then roll over around a week before expiry date?
3. Is there a way to automatically roll over the futures contract, instead of having to manually do it quarterly? I see that Interactive Brokers has an "Auto-roll feature in this link: https://www.interactivebrokers.com/en/s ... llover.htm", but it seems to be only for futures data? Does it auto-roll for my futures position too?
I'm still figuring out some of the logistics myself, so take these answers with a small grain of salt:
  • In HFEA, rebalancing quarterly had the best performance. You need to roll over futures once/quarter anyway, so may as well do your rebalancing then. From a fundamental/conceptual POV, I don't think it matters that much.
  • You don't have to, but that's when volume is highest, so spreads should be lowest. Seems like the best way to go.
  • I was very confused by that as well, but as far as I can tell, that "auto-roll" is only for data purposes, not for execution. So yes, you need to roll manually (sell current contracts, buy contracts for next quarter).
One other point: did you calculate how much collateral you need for those futures positions? At a quick glance, $67k seems like way more than you need, so I would swap some of that MES for a domestic equity ETF of your choice. Reason being: MES has an implied financing cost (something like 0.4% IIRC), whereas the ETF will only have its expense ratio.
Great, in that case, I'll schedule my rebalancing & roll-over quarterly during the highest-volume period.

I didn't have a fixed formula to calculate how much collateral I need, but I configured it on the higher side to provide for more cash buffer. But good point, VTI's ER is only 0.03%, so it makes sense to have more of it.
Bread Investor
Posts: 7
Joined: Sun Nov 07, 2021 10:44 pm

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by Bread Investor »

skierincolorado wrote: Sat Nov 13, 2021 10:39 am
Bread Investor wrote: Sat Nov 13, 2021 12:43 am Thanks Klaus14 and skierincolorado for your inputs. With your inputs, as well as reading through most of the posts in this thread, I've decided on a 165/250 Stock/ITT allocation.

Would the following work for a $100k portfolio?

Image

Some additional questions (I'm using Interactive Brokers) which I need your advice:
1. How often should I do the rebalancing (e.g. Monthly/Quarterly)?
2. For futures (i.e. MES & ZF), is the rule is to always buy the contract that is nearest to expiry, and then roll over around a week before expiry date?
3. Is there a way to automatically roll over the futures contract, instead of having to manually do it quarterly? I see that Interactive Brokers has an "Auto-roll feature in this link: https://www.interactivebrokers.com/en/s ... llover.htm", but it seems to be only for futures data? Does it auto-roll for my futures position too?
A few suggestions: a) do 1 ZF and 1 ZN. The duration of ZF is more like 4.5 years, and the backtests I've posted mostly use ETFs with 5-5.5 year durations. Plus you'll get a little diversification from holding one of each. b) You are near the cutoff for dropping an MES and buying 23k of VTI. You would still hold 44k of cash collateral for 5 MES + 1 ZF + 1 ZN, which is near 7k per contract. c) If you decide to do (b) you could buy VB and/or VBR to make up for the large cap tilt of MES.

Regarding your questions:
1. There has not been much discussion of rebalancing. I would assume similar to HFEA would work best. So far I have just used new funds to rebalance.
2. Yes - although for the treasuries IBKR forces you to roll just over a month before expiry
3. I don't think so. Confusingly that auto-roll is for the data not for the actual investment.
I've adjusted based on your comments, plus a few additional edits ($150k portfolio):

Image

1. Added ZF & ZN. 2ZF + 1ZN (5.4yrs weighted average duration)
2. Added VB
3. Added tech exposure: MNQ (which I assume exposure is quoted price x 2, as per CME's website)
4. Adjusted holdings:
i. Domestic-to-International: 80:20
ii. NonTech-to-Tech: 60:40
5. Adjusted holdings such that collateral can provide for drawdown on 21% Stocks & 8.5% Bonds simultaneously

Image


Would you or anyone else have any opinions/concerns on the above? I'm thinking of getting the above portfolio over the coming 2 weeks
Topic Author
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by skierincolorado »

Bread Investor wrote: Sat Nov 13, 2021 8:28 pm
skierincolorado wrote: Sat Nov 13, 2021 10:39 am
Bread Investor wrote: Sat Nov 13, 2021 12:43 am Thanks Klaus14 and skierincolorado for your inputs. With your inputs, as well as reading through most of the posts in this thread, I've decided on a 165/250 Stock/ITT allocation.

Would the following work for a $100k portfolio?

Image

Some additional questions (I'm using Interactive Brokers) which I need your advice:
1. How often should I do the rebalancing (e.g. Monthly/Quarterly)?
2. For futures (i.e. MES & ZF), is the rule is to always buy the contract that is nearest to expiry, and then roll over around a week before expiry date?
3. Is there a way to automatically roll over the futures contract, instead of having to manually do it quarterly? I see that Interactive Brokers has an "Auto-roll feature in this link: https://www.interactivebrokers.com/en/s ... llover.htm", but it seems to be only for futures data? Does it auto-roll for my futures position too?
A few suggestions: a) do 1 ZF and 1 ZN. The duration of ZF is more like 4.5 years, and the backtests I've posted mostly use ETFs with 5-5.5 year durations. Plus you'll get a little diversification from holding one of each. b) You are near the cutoff for dropping an MES and buying 23k of VTI. You would still hold 44k of cash collateral for 5 MES + 1 ZF + 1 ZN, which is near 7k per contract. c) If you decide to do (b) you could buy VB and/or VBR to make up for the large cap tilt of MES.

Regarding your questions:
1. There has not been much discussion of rebalancing. I would assume similar to HFEA would work best. So far I have just used new funds to rebalance.
2. Yes - although for the treasuries IBKR forces you to roll just over a month before expiry
3. I don't think so. Confusingly that auto-roll is for the data not for the actual investment.
I've adjusted based on your comments, plus a few additional edits ($150k portfolio):

Image

1. Added ZF & ZN. 2ZF + 1ZN (5.4yrs weighted average duration)
2. Added VB
3. Added tech exposure: MNQ (which I assume exposure is quoted price x 2, as per CME's website)
4. Adjusted holdings:
i. Domestic-to-International: 80:20
ii. NonTech-to-Tech: 60:40
5. Adjusted holdings such that collateral can provide for drawdown on 21% Stocks & 8.5% Bonds simultaneously

Image


Would you or anyone else have any opinions/concerns on the above? I'm thinking of getting the above portfolio over the coming 2 weeks
I like all of it except the tech tilt. It's more than 40% tech since most of MES is tech at this point already. All of the other aspects of this strategy are grounded in theoretical concepts with significant academic research. Leveraging while young is part of the Lifecycle Investing framework by Ayres and Nalebuff (and other related research). Diversifying with bonds is supported by MPT. Shorter durations of bonds are supported by the bet against beta. I don't know of a serious theoretical framework for why the outperformance of tech should continue. If anything, the theoretical frameworks I am familiar with would suggest that it is less diversified and sub-optimal - especially when leveraging where low volatility is important.

I'm not familiar with a "tech" factor or a theoretical framework why stocks listed on the Nasdaq instead of other exchanges should outperform in the long-run. In an efficient market there should be no premium for tech or Nasdaq. There is some theoretical support for other factors, such as low-beta, momentum, and SCV (in order of signficance).
keith6014
Posts: 697
Joined: Thu Jan 02, 2014 10:58 pm

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by keith6014 »

This is a naive question. How much cash outlay is needed? I see the cash outlay of $100,000. Is that really required to do a minimal version of the strategy?

I was thinking
1 x MES/Jun2022 - $4663
1 * FV/Feb 2022 - $120'28. FV is 5 year note.

This gives you total notional of ~143k ($23k MES, ~120k ITT). Thats close to 1:8. . Assuming a drawdown of 20% for MES and 10% for ITT, I would need ~$16k in cash collateral.

How wrong am I?
DMoogle
Posts: 549
Joined: Sat Oct 31, 2020 10:24 pm

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by DMoogle »

keith6014 wrote: Sun Nov 14, 2021 6:50 am This is a naive question. How much cash outlay is needed? I see the cash outlay of $100,000. Is that really required to do a minimal version of the strategy?

I was thinking
1 x MES/Jun2022 - $4663
1 * FV/Feb 2022 - $120'28. FV is 5 year note.

This gives you total notional of ~143k ($23k MES, ~120k ITT). Thats close to 1:8. . Assuming a drawdown of 20% for MES and 10% for ITT, I would need ~$16k in cash collateral.

How wrong am I?
How do you define "minimal version of the strategy"? If you don't have the capital to use futures, then just go with the the LETFs in the original HFEA until you do. If you overweight either equities or treasuries, then you hurt the performance (both risk and reward) of the portfolio.

Margin requirements are on CME Group's website. I found a summary here: https://www.tradestation.com/pricing/fu ... uirements/. Spoiler: they require far less cash collateral than you might think.

Problem is, your risk obviously skyrockets. Bogleheads typically aren't interested in "rags to riches or bust" strategies, and that's not what this strategy is nor what it's designed for. I say I have a very high risk tolerance... my total portfolio is a little under $1M, and a sizable chunk of it is in these strategies, but that's because I'm OK if there's a risk that it drops to $100k (90% drawdown), which would be worse than the long-term backtests for would suggest, but I think it makes sense to scale up max drawdown in backtests. Historically, equity crashes have nearly always been cushioned by treasury crashes, but there's still potential for a disaster scenario in which they both crash hard.

I would highly suggest playing around with Portfolio Visualizer to see how long an extremely highly leveraged strategy would really last.
Topic Author
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by skierincolorado »

keith6014 wrote: Sun Nov 14, 2021 6:50 am This is a naive question. How much cash outlay is needed? I see the cash outlay of $100,000. Is that really required to do a minimal version of the strategy?

I was thinking
1 x MES/Jun2022 - $4663
1 * FV/Feb 2022 - $120'28. FV is 5 year note.

This gives you total notional of ~143k ($23k MES, ~120k ITT). Thats close to 1:8. . Assuming a drawdown of 20% for MES and 10% for ITT, I would need ~$16k in cash collateral.

How wrong am I?
I assume by cash outlay, you mean liquid net worth. One should always think of one's investments holistically. I would say you need a bit more liquid net worth than 16k to buy 1 MES and 1 ZF. It would be too much leveage in bonds with only 16k. It would be a 140/750 AA. A dradown in bonds could force you to liquidate everything. Also if you are comfortable with that much risk you'd probably want more stock and less bonds... 140/750 is a pretty unbalanced AA. You'd need other investments so you could target a more balanced AA like 150/250. You'll see people posting in this thread on their total AA across their net worth.. usually numbers like 130/180, 140/140, or 165/250. You need about 30-40k liquid net worth to get into ZF futures, because at 40k liquid net worth 1 ZF will be a 300% 4.5 year duration bond allocation. With less than 30-40k liquid net worth, one could look at LETFs like TYA.

In terms of how much cash you hold as part of a larger portfolio in order to maintain 1 MES + 1 ZF, yeah 16k sounds about right. Personally I hold a bit less than that per MES and per ZF. But when markets drop you sell other assets to maintain the cash collateral (and possibly even buy more MES and ZF).
keith6014
Posts: 697
Joined: Thu Jan 02, 2014 10:58 pm

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by keith6014 »

DMoogle wrote: Sun Nov 14, 2021 9:34 am
keith6014 wrote: Sun Nov 14, 2021 6:50 am This is a naive question. How much cash outlay is needed? I see the cash outlay of $100,000. Is that really required to do a minimal version of the strategy?

I was thinking
1 x MES/Jun2022 - $4663
1 * FV/Feb 2022 - $120'28. FV is 5 year note.

This gives you total notional of ~143k ($23k MES, ~120k ITT). Thats close to 1:8. . Assuming a drawdown of 20% for MES and 10% for ITT, I would need ~$16k in cash collateral.

How wrong am I?
How do you define "minimal version of the strategy"? If you don't have the capital to use futures, then just go with the the LETFs in the original HFEA until you do. If you overweight either equities or treasuries, then you hurt the performance (both risk and reward) of the portfolio.

Margin requirements are on CME Group's website. I found a summary here: https://www.tradestation.com/pricing/fu ... uirements/. Spoiler: they require far less cash collateral than you might think.

Problem is, your risk obviously skyrockets. Bogleheads typically aren't interested in "rags to riches or bust" strategies, and that's not what this strategy is nor what it's designed for. I say I have a very high risk tolerance... my total portfolio is a little under $1M, and a sizable chunk of it is in these strategies, but that's because I'm OK if there's a risk that it drops to $100k (90% drawdown), which would be worse than the long-term backtests for would suggest, but I think it makes sense to scale up max drawdown in backtests. Historically, equity crashes have nearly always been cushioned by treasury crashes, but there's still potential for a disaster scenario in which they both crash hard.

I would highly suggest playing around with Portfolio Visualizer to see how long an extremely highly leveraged strategy would really last.
I define minimal version of strategy as: lowest amount of money needed to test it out. Similar to 55%:45% LETFs with quarterly rebalancing. (ie split $1000: $450 into TMF and the reminder in UPRO). As you I have high risk tolerance but I have little to no experience with futures. I had no experience with LETFs but it was easy to test because such small capital requirements. And now, I only stick with LETFs and liquidate them using covered calls for rebalancing. I guess I need smaller denomination contracts. Something like this will in futures spacehttps://www.prnewswire.com/news-release ... 09732.html.
Hfearless
Posts: 135
Joined: Sat Aug 14, 2021 9:53 am

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by Hfearless »

keith6014 wrote: Sun Nov 14, 2021 10:03 amI guess I need smaller denomination contracts. Something like this will in futures spacehttps://www.prnewswire.com/news-release ... 09732.html.
Micro Treasury Yield futures exist. Proponents of mHFEA don’t seem to have reached a consensus as to how suitable these are for the strategy.
DMoogle
Posts: 549
Joined: Sat Oct 31, 2020 10:24 pm

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by DMoogle »

keith6014 wrote: Sun Nov 14, 2021 10:03 amI define minimal version of strategy as: lowest amount of money needed to test it out. Similar to 55%:45% LETFs with quarterly rebalancing. (ie split $1000: $450 into TMF and the reminder in UPRO). As you I have high risk tolerance but I have little to no experience with futures. I had no experience with LETFs but it was easy to test because such small capital requirements. And now, I only stick with LETFs and liquidate them using covered calls for rebalancing. I guess I need smaller denomination contracts. Something like this will in futures spacehttps://www.prnewswire.com/news-release ... 09732.html.
When you say "test it out," do you mean you're looking for ways to familiarize yourself with the mechanics of managing the account? Could just open a paper trading account and play with that before using real money.
Hfearless wrote: Sun Nov 14, 2021 10:32 am
keith6014 wrote: Sun Nov 14, 2021 10:03 amI guess I need smaller denomination contracts. Something like this will in futures spacehttps://www.prnewswire.com/news-release ... 09732.html.
Micro Treasury Yield futures exist. Proponents of mHFEA don’t seem to have reached a consensus as to how suitable these are for the strategy.
Huh, I didn't know that. Looks like they launched just a few months ago - last June. They should probably be at least considered for individuals with <$100k of capital.

At a glance, however, I see two downsides:
  • It's far lower volume than the non-micro treasuries, i.e. probably bigger spreads/less efficiency (although maybe this isn't that big of a deal).
  • The expiration schedule is monthly, not quarterly. Since practically all the volume is with the nearest month, this means you'll have to roll over monthly. Little bit of a hassle.
Interesting to hear they're launching a "nano" size contract. Curious to see if that takes off.
Hfearless
Posts: 135
Joined: Sat Aug 14, 2021 9:53 am

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by Hfearless »

Perhaps this thread is indeed the better place to ask this question. On one hand, futures are supposed to track their underlying very closely. On the other hand, there are periods such as 2016–2019 when bond ETFs such as VGIT, IEF, SPTI grow (+2.5…3% for the entire period) but /ZF and /ZN fall about 4% according to Yahoo Finance data. Why should that happen?
keith6014
Posts: 697
Joined: Thu Jan 02, 2014 10:58 pm

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by keith6014 »

DMoogle wrote: Sun Nov 14, 2021 10:58 am
keith6014 wrote: Sun Nov 14, 2021 10:03 amI define minimal version of strategy as: lowest amount of money needed to test it out. Similar to 55%:45% LETFs with quarterly rebalancing. (ie split $1000: $450 into TMF and the reminder in UPRO). As you I have high risk tolerance but I have little to no experience with futures. I had no experience with LETFs but it was easy to test because such small capital requirements. And now, I only stick with LETFs and liquidate them using covered calls for rebalancing. I guess I need smaller denomination contracts. Something like this will in futures spacehttps://www.prnewswire.com/news-release ... 09732.html.
When you say "test it out," do you mean you're looking for ways to familiarize yourself with the mechanics of managing the account? Could just open a paper trading account and play with that before using real money.
Hfearless wrote: Sun Nov 14, 2021 10:32 am
keith6014 wrote: Sun Nov 14, 2021 10:03 amI guess I need smaller denomination contracts. Something like this will in futures spacehttps://www.prnewswire.com/news-release ... 09732.html.
Micro Treasury Yield futures exist. Proponents of mHFEA don’t seem to have reached a consensus as to how suitable these are for the strategy.
Huh, I didn't know that. Looks like they launched just a few months ago - last June. They should probably be at least considered for individuals with <$100k of capital.

At a glance, however, I see two downsides:
  • It's far lower volume than the non-micro treasuries, i.e. probably bigger spreads/less efficiency (although maybe this isn't that big of a deal).
  • The expiration schedule is monthly, not quarterly. Since practically all the volume is with the nearest month, this means you'll have to roll over monthly. Little bit of a hassle.
Interesting to hear they're launching a "nano" size contract. Curious to see if that takes off.
Testing it meaning, trying with small amount of money. I don't know if Schwab has a paper trading feature. Yeah, I am really looking forward to the nano size contracts. The last 10 years things have been getting easier for retail/amateur investors. Not sure if thats a good thing but I like it. Thanks again for your kind response and suggestions!
msun641
Posts: 14
Joined: Sun Jan 24, 2021 1:36 pm

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by msun641 »

Hfearless wrote: Sun Nov 14, 2021 11:37 am Perhaps this thread is indeed the better place to ask this question. On one hand, futures are supposed to track their underlying very closely. On the other hand, there are periods such as 2016–2019 when bond ETFs such as VGIT, IEF, SPTI grow (+2.5…3% for the entire period) but /ZF and /ZN fall about 4% according to Yahoo Finance data. Why should that happen?
My own (cursory) understanding is that futures throw off more of the implied yield through the cash-settled component, making the price over longer periods of time more directly tied to the interest rate only. If you were to track the price of ZN over the 3-month life of a single contract, you would see the yield component more clearly.

Please someone correct me/explain it better than I can!
Topic Author
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by skierincolorado »

Hfearless wrote: Sun Nov 14, 2021 11:37 am Perhaps this thread is indeed the better place to ask this question. On one hand, futures are supposed to track their underlying very closely. On the other hand, there are periods such as 2016–2019 when bond ETFs such as VGIT, IEF, SPTI grow (+2.5…3% for the entire period) but /ZF and /ZN fall about 4% according to Yahoo Finance data. Why should that happen?
msun641 wrote: Sun Nov 14, 2021 6:47 pm

My own (cursory) understanding is that futures throw off more of the implied yield through the cash-settled component, making the price over longer periods of time more directly tied to the interest rate only. If you were to track the price of ZN over the 3-month life of a single contract, you would see the yield component more clearly.

Please someone correct me/explain it better than I can!
I don't think this is correct. Futures reflect all of the return of the underlying, the same as an ETF. Upthread there are some graphs of futures returns vs ETF returns and they are identical, when matched to the appropriate duration.

The reason that ZF or ZN aren't matching is most likely that they are not the same duration as the ETF they are being compared to, or a mistake is being made. Perhaps the yahoo futures return is not a total return index and is a futures price index. There are some graphs upthread of this or similar comparisons.
Bread Investor
Posts: 7
Joined: Sun Nov 07, 2021 10:44 pm

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by Bread Investor »

Thanks skierincolorado.

For a 165:250 portfolio, if S&P500 drops by 33%, am I right to say that the Stocks component will be fully liquidated? In that event, I'll then need to sell bonds to maintain the 165:250 portfolio?
Hfearless
Posts: 135
Joined: Sat Aug 14, 2021 9:53 am

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by Hfearless »

Bread Investor wrote: Sun Nov 14, 2021 7:53 pm For a 165:250 portfolio, if S&P500 drops by 33%, am I right to say that the Stocks component will be fully liquidated?
No. Only a 300% allocation loses all its value upon a 33% drop of the underlying. If you implemented your 1.65x leverage using a 3x leveraged ETF, you might expect such a LETF to be annihilated in such an event, but the quirks of daily rebalancing sometimes work in your favor:

2020-02-10: SPY=337.6, UPRO=79.86
2020-03-16: SPY=228.8 (−32%), UPRO=20.30 (−75%)

To destroy UPRO, SPX has to drop 33% in a single day, except this can’t happen because a 20% drop engages the level 3 circuit breaker, halting trading for the remainder of the day.
User avatar
Bentonkb
Posts: 83
Joined: Sun Mar 07, 2021 5:47 am

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by Bentonkb »

keith6014 wrote: Sun Nov 14, 2021 6:14 pm
DMoogle wrote: Sun Nov 14, 2021 10:58 am
keith6014 wrote: Sun Nov 14, 2021 10:03 amI define minimal version of strategy as: lowest amount of money needed to test it out. Similar to 55%:45% LETFs with quarterly rebalancing. (ie split $1000: $450 into TMF and the reminder in UPRO). As you I have high risk tolerance but I have little to no experience with futures. I had no experience with LETFs but it was easy to test because such small capital requirements. And now, I only stick with LETFs and liquidate them using covered calls for rebalancing. I guess I need smaller denomination contracts. Something like this will in futures spacehttps://www.prnewswire.com/news-release ... 09732.html.
When you say "test it out," do you mean you're looking for ways to familiarize yourself with the mechanics of managing the account? Could just open a paper trading account and play with that before using real money.
Hfearless wrote: Sun Nov 14, 2021 10:32 am
keith6014 wrote: Sun Nov 14, 2021 10:03 amI guess I need smaller denomination contracts. Something like this will in futures spacehttps://www.prnewswire.com/news-release ... 09732.html.
Micro Treasury Yield futures exist. Proponents of mHFEA don’t seem to have reached a consensus as to how suitable these are for the strategy.
Huh, I didn't know that. Looks like they launched just a few months ago - last June. They should probably be at least considered for individuals with <$100k of capital.

At a glance, however, I see two downsides:
  • It's far lower volume than the non-micro treasuries, i.e. probably bigger spreads/less efficiency (although maybe this isn't that big of a deal).
  • The expiration schedule is monthly, not quarterly. Since practically all the volume is with the nearest month, this means you'll have to roll over monthly. Little bit of a hassle.
Interesting to hear they're launching a "nano" size contract. Curious to see if that takes off.
Testing it meaning, trying with small amount of money. I don't know if Schwab has a paper trading feature. Yeah, I am really looking forward to the nano size contracts. The last 10 years things have been getting easier for retail/amateur investors. Not sure if thats a good thing but I like it. Thanks again for your kind response and suggestions!
Don't get too excited about the new micro treasuries on the Small Exchange. They are directly tied to the interest rates, rather than a basket of bonds, so they don't have any expected return from a coupon payment. I don't think there is any roll down yield either, because the index is tracking the yield at a fixed spot on the yield curve.

There might be more to it if you look at the appropriate no-arb conditions, just like skier and company have done in this thread for the treasury futures. My simplistic understanding of the way the new micros work does seem to create an arb opportunity. You could go long on one of the ultra treasury futures (like TN) that has a narrow range of possible durations and also go long on the matching micro (S10Y) to hedge against interest rate changes. The result would be a risk free coupon yield from the TN.

Before you try to use S10Y and the other offerings from the Small Exchange, you better make sure that the hedge funds are doing their jobs and eliminating the arb opportunity.
Bread Investor
Posts: 7
Joined: Sun Nov 07, 2021 10:44 pm

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by Bread Investor »

Hfearless wrote: Mon Nov 15, 2021 6:03 am
Bread Investor wrote: Sun Nov 14, 2021 7:53 pm For a 165:250 portfolio, if S&P500 drops by 33%, am I right to say that the Stocks component will be fully liquidated?
No. Only a 300% allocation loses all its value upon a 33% drop of the underlying. If you implemented your 1.65x leverage using a 3x leveraged ETF, you might expect such a LETF to be annihilated in such an event, but the quirks of daily rebalancing sometimes work in your favor:

2020-02-10: SPY=337.6, UPRO=79.86
2020-03-16: SPY=228.8 (−32%), UPRO=20.30 (−75%)

To destroy UPRO, SPX has to drop 33% in a single day, except this can’t happen because a 20% drop engages the level 3 circuit breaker, halting trading for the remainder of the day.
Interesting. Taking the two time periods that you've quoted, the 3x LEFT would have outperformed a similarly leveraged position using S&P Futures? This is since when SPY dropped 32%, the Futures position would have dropped 96%, while the LEFT only dropped 75%.
DMoogle
Posts: 549
Joined: Sat Oct 31, 2020 10:24 pm

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by DMoogle »

Bread Investor wrote: Mon Nov 15, 2021 7:37 amInteresting. Taking the two time periods that you've quoted, the 3x LEFT would have outperformed a similarly leveraged position using S&P Futures? This is since when SPY dropped 32%, the Futures position would have dropped 96%, while the LEFT only dropped 75%.
Yes. It's a function of the daily rebalancing. Think of it this way:

If the index drops 10% one day, then another 10% the next, how will the leveraged portfolio react?

In a 2x LETF with $100, you'd drop to ($100-2*10%*$100)=$80 after the first day, then ($80-2*10%*$80)=$64 after the second day.

In a 2x leveraged portfolio that didn't rebalance, you maintain your $100 in exposure after the first day (you still own the same futures contracts or the same # of shares of stock). So, you'd drop to the same ($100-2*10%*$100)=$80 after the first day, then ($80-2*10%*$100)=$60 after the second day.

Basically, LETFs buy high and sell low, which ends up being a good thing if the market continues to soar or plummet (but will get wrecked in "roller coaster" markets).

EDIT: My math is off - see post directly below mine.
Last edited by DMoogle on Mon Nov 15, 2021 8:37 am, edited 1 time in total.
Hfearless
Posts: 135
Joined: Sat Aug 14, 2021 9:53 am

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by Hfearless »

DMoogle wrote: Mon Nov 15, 2021 7:57 am you maintain your $100 in exposure after the first day
That’s incorrect. Your exposure is the current market value of your position. If your futures contract, notional value $100k, dropped 99% to $1k, you lost $99k (ouch). If on the next day it drops 25% to $750, you lose a further $250, not $25k.

What you maintain is your debt if you bought your securities on margin, but that’s not particularly important here.

Underlying:

Code: Select all

the day before yesterday $100
fell 10% yesterday       $90
fell 10% today           $81
total loss               $19
Leveraged 2x using debt or futures or whatever:

Code: Select all

the day before yesterday $200 of exposure, NAV $100, leverage 2x
fell 10% yesterday       $180 of exposure, NAV $80, leverage becomes 2.25x
fell 10% today           $162
total loss               $38 (= twice the loss of the underlying)
2x LETF:

Code: Select all

the day before yesterday $100
fell 10% yesterday       $80 = $100×(1−20%). Leverage reset to 2x.
fell 10% today           $64 = $80×(1−20%)
total loss               $36 (slightly less bad than twice the loss)
DMoogle
Posts: 549
Joined: Sat Oct 31, 2020 10:24 pm

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by DMoogle »

My bad, you're correct, that's what I get for doing napkin math first thing in the morning. :sharebeer
Hfearless
Posts: 135
Joined: Sat Aug 14, 2021 9:53 am

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by Hfearless »

Bentonkb wrote: Mon Nov 15, 2021 7:22 am Don't get too excited about the new micro treasuries on the Small Exchange. They are directly tied to the interest rates, rather than a basket of bonds, so they don't have any expected return from a coupon payment. I don't think there is any roll down yield either, because the index is tracking the yield at a fixed spot on the yield curve.
Can we settle this once and for all—do big traditional futures have any expected return?

My understanding is that the total return of a futures contract is the movement of its price plus the profit that can be made from investing the capital in the money market, given that the contract itself doesn’t take up any capital. When implementing (m)HFEA using futures, the return is basically the ~1% that we don’t have to pay as the LETF expense ratio or the loan rate, because we have all that capital with which to buy shares at a low leverage. Which seems to suggest that once rates rise a percentage point or two, LETFs will suddenly become more favorable unless their ERs rise, too. Or am I severely mistaken?
Booogle
Posts: 605
Joined: Wed Nov 09, 2016 2:57 pm

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by Booogle »

How about 135% NTSX?

100% NTSX

35% more on margin.

1.0% interest rate.
Hfearless
Posts: 135
Joined: Sat Aug 14, 2021 9:53 am

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by Hfearless »

Booogle wrote: Mon Nov 15, 2021 8:48 am How about 135% NTSX?
Not bad, but you can do better. 135% NTSX = 120% SPY + 80% IEF, research in this thread suggests more bonds than that.
User avatar
Bentonkb
Posts: 83
Joined: Sun Mar 07, 2021 5:47 am

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by Bentonkb »

Hfearless wrote: Mon Nov 15, 2021 8:44 am
Bentonkb wrote: Mon Nov 15, 2021 7:22 am Don't get too excited about the new micro treasuries on the Small Exchange. They are directly tied to the interest rates, rather than a basket of bonds, so they don't have any expected return from a coupon payment. I don't think there is any roll down yield either, because the index is tracking the yield at a fixed spot on the yield curve.
Can we settle this once and for all—do big traditional futures have any expected return?

My understanding is that the total return of a futures contract is the movement of its price plus the profit that can be made from investing the capital in the money market, given that the contract itself doesn’t take up any capital. When implementing (m)HFEA using futures, the return is basically the ~1% that we don’t have to pay as the LETF expense ratio or the loan rate, because we have all that capital with which to buy shares at a low leverage. Which seems to suggest that once rates rise a percentage point or two, LETFs will suddenly become more favorable unless their ERs rise, too. Or am I severely mistaken?
Yes, the traditional futures (ZF, ZN, ZB, etc) have an expected return. Some of the return comes from the coupon payment from the underlying bond. Some comes from the roll down as the contract matures and moves to a shorter duration on the yield curve. There is some volatility on top of that due to changes in interest rates, which is the answer to your second question. When rates are rising you might lose more due to the volatility than you make due to yield and roll. I don't think that LETFs would necessarily become more favorable when rates rise.
Hfearless
Posts: 135
Joined: Sat Aug 14, 2021 9:53 am

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by Hfearless »

Bentonkb wrote: Mon Nov 15, 2021 9:46 am Yes, the traditional futures (ZF, ZN, ZB, etc) have an expected return. Some of the return comes from the coupon payment from the underlying bond. Some comes from the roll down as the contract matures and moves to a shorter duration on the yield curve.
Could you please ELI5 how exactly does that get priced in? Is it not true that today’s price is the market’s best estimate of what the market price of the CTD bond is going to be on Dec 31? Which is wholly defined by estimates of what will happen after that date and anything the bond might yield before that does not matter, that payment is collected by whoever owns the bond? Not to mention most holders of futures contracts don’t own any bonds? And even if they do, some people that might use /ZB to sell their bonds hold bonds issued anywhere between 1996 and 2006, with vastly different coupon rates? The futures are standardized using a hypothetical 6% bond, certainly there’s no 6% return on futures?

Let’s compare this to commodity futures. Bonds yield money when held, goods tend to require money for storage and so can be said to have negative yields. But someone with a long position in such a contract most certainly doesn’t incur such loss?
Topic Author
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by skierincolorado »

Hfearless wrote: Mon Nov 15, 2021 8:44 am
Bentonkb wrote: Mon Nov 15, 2021 7:22 am Don't get too excited about the new micro treasuries on the Small Exchange. They are directly tied to the interest rates, rather than a basket of bonds, so they don't have any expected return from a coupon payment. I don't think there is any roll down yield either, because the index is tracking the yield at a fixed spot on the yield curve.
Can we settle this once and for all—do big traditional futures have any expected return?

My understanding is that the total return of a futures contract is the movement of its price plus the profit that can be made from investing the capital in the money market, given that the contract itself doesn’t take up any capital. When implementing (m)HFEA using futures, the return is basically the ~1% that we don’t have to pay as the LETF expense ratio or the loan rate, because we have all that capital with which to buy shares at a low leverage. Which seems to suggest that once rates rise a percentage point or two, LETFs will suddenly become more favorable unless their ERs rise, too. Or am I severely mistaken?
The return of futures is identical to the underlying bonds or an etf of the underlying bonds. This was proved definitely upthread with multiple lines of proof (theoretical and empirical). There is a small cost of around .2% similar to an etf fee. Futures are strictly superior to LETFs unless the cost rises.
User avatar
Bentonkb
Posts: 83
Joined: Sun Mar 07, 2021 5:47 am

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by Bentonkb »

Hfearless wrote: Mon Nov 15, 2021 10:20 am
Bentonkb wrote: Mon Nov 15, 2021 9:46 am Yes, the traditional futures (ZF, ZN, ZB, etc) have an expected return. Some of the return comes from the coupon payment from the underlying bond. Some comes from the roll down as the contract matures and moves to a shorter duration on the yield curve.
Could you please ELI5 how exactly does that get priced in? Is it not true that today’s price is the market’s best estimate of what the market price of the CTD bond is going to be on Dec 31? Which is wholly defined by estimates of what will happen after that date and anything the bond might yield before that does not matter, that payment is collected by whoever owns the bond? Not to mention most holders of futures contracts don’t own any bonds? And even if they do, some people that might use /ZB to sell their bonds hold bonds issued anywhere between 1996 and 2006, with vastly different coupon rates? The futures are standardized using a hypothetical 6% bond, certainly there’s no 6% return on futures?

Let’s compare this to commodity futures. Bonds yield money when held, goods tend to require money for storage and so can be said to have negative yields. But someone with a long position in such a contract most certainly doesn’t incur such loss?
Someone with a long position in, say, gold or oil does incur a loss associated with the storage cost. The second contract will be more expensive than the front contract when you do the quarterly roll. That is how you compensate the counterparty who is holding physical gold while they are waiting for your long position to mature.

The futures have some extra considerations, though. There is a basket of acceptable bonds for each contract and the short gets to choose which one to deliver, so the long position gets a small option premium associated with the long option that the short futures position has in it. The short position gets the coupon payment, so the long position trades at a slight discount to balance out the portion of the coupon that will be paid during the contract. There is also some shift in duration during the contract period, so there is normally some yield from that too.

All that needs to be baked into the quoted price of the bond future contract so that there isn't a risk free arbitrage available.

We had a long discussion of how the 6% reference bond yield affects the futures. It is very confusing. Now that I understand it better, I think of the TCF (Treasury Conversion Factor - the fudge factor that prices all the bonds as if they yielded 6%) as a type of leverage ratio. It is a multiplier that increases (or decreases) the volatility of the cheapest to deliver bond to match that of the contract.
Hfearless
Posts: 135
Joined: Sat Aug 14, 2021 9:53 am

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by Hfearless »

Bentonkb wrote: Mon Nov 15, 2021 11:13 am Someone with a long position in, say, gold or oil does incur a loss associated with the storage cost. The second contract will be more expensive than the front contract when you do the quarterly roll. That is how you compensate the counterparty who is holding physical gold while they are waiting for your long position to mature.
Are you sure? Imagine a commodity that’s only produced in summer and only useful in winter, maybe a food product. Its storage cost will be included in the total price when sold. The market knows this perfectly well, so this will already be accounted for in the contract price. So all costs will be borne by the consumer of the good, not by any holder of a futures contract (of which there could be much more than the total amount of the physical commodity in existence).

Let’s say a unit of it typically costs $100k to produce and $20k to store for half a year. The Sep 30 future would be priced at $110k and Dec 31, $120k. When rolling, yes, prices will be different, but should it turn out that the costs coincided exactly with what the market predicted, no holders of any future contracts will gain or lose anything. Finally, the speculators will close their positions, the real buyers will take delivery and the real sellers will get their $120k.
Topic Author
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by skierincolorado »

Hfearless wrote: Mon Nov 15, 2021 11:38 am
Bentonkb wrote: Mon Nov 15, 2021 11:13 am Someone with a long position in, say, gold or oil does incur a loss associated with the storage cost. The second contract will be more expensive than the front contract when you do the quarterly roll. That is how you compensate the counterparty who is holding physical gold while they are waiting for your long position to mature.
Are you sure? Imagine a commodity that’s only produced in summer and only useful in winter, maybe a food product. Its storage cost will be included in the total price when sold. The market knows this perfectly well, so this will already be accounted for in the contract price. So all costs will be borne by the consumer of the good, not by any holder of a futures contract (of which there could be much more than the total amount of the physical commodity in existence).

Let’s say a unit of it typically costs $100k to produce and $20k to store for half a year. The Sep 30 future would be priced at $110k and Dec 31, $120k. When rolling, yes, prices will be different, but should it turn out that the costs coincided exactly with what the market predicted, no holders of any future contracts will gain or lose anything. Finally, the speculators will close their positions, the real buyers will take delivery and the real sellers will get their $120k.
I think you said the same thing as him. The next month is more expensive than the front month - the same as in your example.

The situation with futures is the same but with negative storage cost - the coupon. The contract price will be cheaper the farther from expiration - otherwise an arb opportunity exists.
Hfearless
Posts: 135
Joined: Sat Aug 14, 2021 9:53 am

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by Hfearless »

skierincolorado wrote: Mon Nov 15, 2021 12:31 pm The contract price will be cheaper the farther from expiration - otherwise an arb opportunity exists.
For the same expiration date of the future? Doesn’t that offer an arbitrage opportunity if I hedge away the interest rate risk using the micro futures?
Topic Author
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by skierincolorado »

Hfearless wrote: Mon Nov 15, 2021 1:11 pm
skierincolorado wrote: Mon Nov 15, 2021 12:31 pm The contract price will be cheaper the farther from expiration - otherwise an arb opportunity exists.
For the same expiration date of the future? Doesn’t that offer an arbitrage opportunity if I hedge away the interest rate risk using the micro futures?
Comeinvest convinced me the micros must also reflect the return of the underlying or else arb would exist. I haven’t investigated further to confirm, but this absolutely must be the case.
Hfearless
Posts: 135
Joined: Sat Aug 14, 2021 9:53 am

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by Hfearless »

What about neither caring about the return, only about the difference between the predicted price and the actual price?

https://imgur.com/a/h8XFSS7

Here ZN and VFITX match each other very closely. From July 2006 to July 2007, a period where the 10y notes consistently yielded about 5%, both did not change their prices—except VFITX did distribute 5% in dividends!
skierincolorado wrote: Mon Nov 15, 2021 1:41 pm Comeinvest convinced me the micros must also reflect the return of the underlying or else arb would exist. I haven’t investigated further to confirm, but this absolutely must be the case.
Also, what exactly the underlying is? The hypothetical 6% bond?
Topic Author
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by skierincolorado »

Hfearless wrote: Mon Nov 15, 2021 2:19 pm What about neither caring about the return, only about the difference between the predicted price and the actual price?

https://imgur.com/a/h8XFSS7

Here ZN and VFITX match each other very closely. From July 2006 to July 2007, a period where the 10y notes consistently yielded about 5%, both did not change their prices—except VFITX did distribute 5% in dividends!
skierincolorado wrote: Mon Nov 15, 2021 1:41 pm Comeinvest convinced me the micros must also reflect the return of the underlying or else arb would exist. I haven’t investigated further to confirm, but this absolutely must be the case.
Also, what exactly the underlying is? The hypothetical 6% bond?
You must have missed it... the yahoo data is just futures price... does not include the return.

the underlying is the CTD bond
Hfearless
Posts: 135
Joined: Sat Aug 14, 2021 9:53 am

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by Hfearless »

skierincolorado wrote: Mon Nov 15, 2021 2:27 pm You must have missed it... the yahoo data is just futures price... does not include the return.
the underlying is the CTD bond
Are you referring to the return that I would get if I opened a futures position and invested capital equal to its notional value in the money market?
Topic Author
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: Modified versions of HFEA with ITT and Futures / Lifecycle Investing with Modern Portfolio Theory

Post by skierincolorado »

Hfearless wrote: Mon Nov 15, 2021 2:29 pm
skierincolorado wrote: Mon Nov 15, 2021 2:27 pm You must have missed it... the yahoo data is just futures price... does not include the return.
the underlying is the CTD bond
Are you referring to the return that I would get if I opened a futures position and invested capital equal to its notional value in the money market?
Not only that, but also the return you'd get from the price increases that occur as futures contracts approach expiry.

The only thing that the yahoo treasury futures price index is telling you is the history of prices. So for example, hypothetically it could stay at 120,000 for ZF forever. That would suggest there was no return. But in reality each contract was bought at 119,000 and sold for 120,000. That's a $1000 profit every three months. Yahoo splices all of these together and you just see a constant price of ~120,000. So you see a flat line suggesting no return. In reality you were getting $4000/ yr or nearly 4%. Plus the cash return on the capital you freed up.

You need to use a total return index like the SPGlobal data.

For example, Yahoo shows the "price" of ZF going from 118 to 121 in the last 5 years. (less than 3% increase). The only thing this reflects is a shift in the conversion factor. It actually doesn't reflect return at all. It's not even related to return.. it's a completely useless metric.

https://www.spglobal.com/spdji/en/indic ... /#overview

SPGlobal shows the total return index going from 542 to 596 (10% return)

https://www.spglobal.com/spdji/en/indic ... /#overview


This is all covered extensively upthread :D
Post Reply