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.
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 »

Chocolatebar wrote: Mon Mar 06, 2023 2:41 pm . My logic for 90/10 stock/bonds is basically that since the market is currently down, I don't need as many bonds,
Since you posted this 2 year interest rates fell from a high of 5.1% to a brief low of 3.8%. Currently at 4.35%. Huge money to be made. Stocks down. My IB account is up 20%. I don't mean to troll, just to make the point that the market works in unexpected ways. Don't try to predict it. Things could go either way from here. Maybe this is just the beginning of banking sector implosion and rate decreases, maybe things settle down. Again I don't think there is a problem with holding limited or no bonds, but there could be a problem with an attempt to dynamically time them. Not suggesting you add more bonds at this point, unless its part of a long term plan. Just the big picture of market timing.
Chocolatebar
Posts: 192
Joined: Tue Feb 14, 2023 10:58 am

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

Post by Chocolatebar »

skierincolorado wrote: Tue Mar 14, 2023 9:48 am Huge money to be made.
What do you mean?
skierincolorado wrote: Tue Mar 14, 2023 9:48 am Stocks down. My IB account is up 20%. I don't mean to troll, just to make the point that the market works in unexpected ways. Don't try to predict it.
How am I doing that?
skierincolorado wrote: Tue Mar 14, 2023 9:48 am Things could go either way from here. Maybe this is just the beginning of banking sector implosion and rate decreases, maybe things settle down. Again I don't think there is a problem with holding limited or no bonds, but there could be a problem with an attempt to dynamically time them. Not suggesting you add more bonds at this point, unless its part of a long term plan. Just the big picture of market timing.
I feel like you missed the post where I explained that I'm not timing my asset allocation. I'm simply adjusting it as my net worth changes, which is what everyone else does. When it increases significantly - I should carry more bonds and less stocks. When it dramatically decreases via market crash - I should load up on stocks.

If you disagree with this then it's totally fine. I concede that everyone else in this thread is more educated than I am when it comes to investing. If you have a recommendation for what I should do instead then I'd love to hear it. The problem with my strategy at this point is that I'm limited until I'm comfortable utilizing futures and box spreads.

BTW - my daughter was born a few days ago and everyone is happy and healthy. Proud papa here :sharebeer
Last edited by Chocolatebar on Tue Mar 14, 2023 3:09 pm, edited 1 time in total.
comeinvest
Posts: 2669
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest »

I'm mildly confused reconciling the data sources of LIBOR and Term SOFR, can somebody please help?
For 03/13, we have SOFR at 4.80 here https://www.traditiondata.com/americas/us-sofr/

Image

and 4.73 here https://www.theice.com/iba/term-rates

Image

But CME says it was 5.03 https://www.cmegroup.com/trading/equity ... /main.html

Image

And ICE has LIBOR at 5.18 https://www.theice.com/pace-of-roll/MSC ... 7-20230616 (although it's 4.87 according to the ICE link in the pic above https://www.theice.com/iba/term-rates)

Image

but it was 4.87 according to here https://www.global-rates.com/en/interes ... libor.aspx

Image

Which one is right? Slightly different reference times perhaps if they don't take end of day values, but still?
comeinvest
Posts: 2669
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest »

skierincolorado wrote: Tue Mar 14, 2023 9:48 am
Chocolatebar wrote: Mon Mar 06, 2023 2:41 pm . My logic for 90/10 stock/bonds is basically that since the market is currently down, I don't need as many bonds,
Since you posted this 2 year interest rates fell from a high of 5.1% to a brief low of 3.8%. Currently at 4.35%. Huge money to be made. Stocks down. My IB account is up 20%. I don't mean to troll, just to make the point that the market works in unexpected ways. Don't try to predict it. Things could go either way from here. Maybe this is just the beginning of banking sector implosion and rate decreases, maybe things settle down. Again I don't think there is a problem with holding limited or no bonds, but there could be a problem with an attempt to dynamically time them. Not suggesting you add more bonds at this point, unless its part of a long term plan. Just the big picture of market timing.
The S&P500 is 2.4% down, VGIT (duration 5.2 years) is 2.2% up over the last 5 days to right now. For whoever had about equal amounts of equities and ITT, the portfolio value didn't change much. Whoever had STT or short maturity SOFR futures in equivalent amounts, had an overall gain. The short-term rates dropped fastest.

I use tax-exempt municipal bond closed-end funds in my taxable account. The index ETFs VTEB and MUB both rose about 1% during the last 5 days. Not quite as fast as treasuries.
I use municipal bonds with long maturities, which have much higher after-tax term premium and yield spread to risk-free rates than treasuries. VWLTX (Vanguard Long-Term Tax-Exempt Fund) and VCITX (Vanguard California Long-Term Tax-Exempt Fund) rose about 1.5% over the last week i.e. from 03/06 to yesterday 03/13. For comparison, VUSTX (Vanguard Long-Term Treasury Fund) rose about 4% in the same timeframe. VGLT (Vanguard Long-Term Treasury Index ETF) rose about 4.5% until yesterday, but dropped 1.5% today, for a total gain of 2% over the last 5 days until now.
The NAV of BFZ and EVM (leveraged California municipal bonds) rose about 2.5% between 03/06 and 03/13.

For comparison, VCLT (Vanguard Long-Term Corporate Bond fund) rose about 0.5% during the last 5 days.

The popular managed futures ETFs KMLM and DBMF which worked so nicely last year when everything else fell, had -3% and -4% performance over the last 5 days, respectively.

But not every crisis is equal. In March 2020, municipal bonds dropped first before they rose, and even treasuries exhibited erratic behavior momentarily, if I remember right.
Last edited by comeinvest on Wed Mar 15, 2023 12:42 am, edited 1 time 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 »

Chocolatebar wrote: Tue Mar 14, 2023 2:04 pm
skierincolorado wrote: Tue Mar 14, 2023 9:48 am Huge money to be made.
What do you mean?
skierincolorado wrote: Tue Mar 14, 2023 9:48 am Stocks down. My IB account is up 20%. I don't mean to troll, just to make the point that the market works in unexpected ways. Don't try to predict it.
How am I doing that?
skierincolorado wrote: Tue Mar 14, 2023 9:48 am Things could go either way from here. Maybe this is just the beginning of banking sector implosion and rate decreases, maybe things settle down. Again I don't think there is a problem with holding limited or no bonds, but there could be a problem with an attempt to dynamically time them. Not suggesting you add more bonds at this point, unless its part of a long term plan. Just the big picture of market timing.
I feel like you missed the post where I explained that I'm not timing my asset allocation. I'm simply adjusting it as my net worth changes, which is what everyone else does. When it increases significantly - I should carry more bonds and less stocks. When it dramatically decreases via market crash - I should load up on stocks.

If you disagree with this then it's totally fine. I concede that everyone else in this thread is more educated than I am when it comes to investing. If you have a recommendation for what I should do instead then I'd love to hear it. The problem with my strategy at this point is that I'm limited until I'm comfortable utilizing futures and box spreads.

BTW - my daughter was born a few days ago and everyone is happy and healthy. Proud papa here :sharebeer
Congratulations! Will read your other post and respond when I can.
comeinvest
Posts: 2669
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest »

I would say the reference interest rates on the CME page are definitely not reliable. The 5.03% 3M SOFR on 03/13 same as on 03/10 doesn't match the SOFR future movement. I think this Mar 2023 SOFR futures contract should closely match the financing period of the CME equity futures expiring in June. Fair enough, they make the SOFR rate field editable; but why don't they put in the correct rate?!

Image

Image
comeinvest
Posts: 2669
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest »

On 03/13 the difference of S&P 400 and S&P 500 implied rate widened again to a relatively typical 0.41%.
It looks like I lucked out and got and additional 8.5 bps with my fills, locking in a 0.495% p.a. spread difference with my offsetting equity futures pair for the next quarter, if my math is right.

Image

Image
comeinvest
Posts: 2669
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest »

On the ICE site it looks like the MXEF (MME) futures roll cheapened on 03/13; but their reference rates are off; perhaps they take beginning of day values? Rates decreased about 0.2% on average vs the prior day, and 3M SOFR was more like 4.8% on average during the day, making the spread to SOFR on 03/13 about 0.9%. I had a standing limit order while rates were crushing, and paid even more. The MSCI EAFE futures roll was about equally expensive. Not fun.

Image
comeinvest
Posts: 2669
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest »

Nikkei futures in JPY provide cheap implied financing at about 0.15% above the risk-free rate / government bonds.

I'm not sure what exactly my exposure to volatilities would be if I were to use USD based quanto Nikkei futures, which might become relevant for collateral management when the JPY short-term interest rates rise measurably above zero.

Image

Image
comeinvest
Posts: 2669
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest »

A fantastic correlation regime change occurred around March 6th. Look at that precision in the second chart.

Image

Image
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 »

comeinvest wrote: Wed Mar 15, 2023 3:14 am On 03/13 the difference of S&P 400 and S&P 500 implied rate widened again to a relatively typical 0.41%.
It looks like I lucked out and got and additional 8.5 bps with my fills, locking in a 0.495% p.a. spread difference with my offsetting equity futures pair for the next quarter, if my math is right.

Image

Image
Did you see they will have a micro mid cap starting March 20th? I am curious what the implied rate will be. If I understand, emd is 250k per contract which I don't quite have enough for. I'm getting close but I'm also hesitant to switch 250k straight from MES to EMD all at once.
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 »

comeinvest wrote: Wed Mar 15, 2023 3:33 am On the ICE site it looks like the MXEF (MME) futures roll cheapened on 03/13; but their reference rates are off; perhaps they take beginning of day values? Rates decreased about 0.2% on average vs the prior day, and 3M SOFR was more like 4.8% on average during the day, making the spread to SOFR on 03/13 about 0.9%. I had a standing limit order while rates were crushing, and paid even more. The MSCI EAFE futures roll was about equally expensive. Not fun.

Image
Wow expensive. Why buy msci futures at all? I might buy just 1 mxef for diversification and dollar hedging.
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 »

comeinvest wrote: Thu Mar 16, 2023 12:36 pm A fantastic correlation regime change occurred around March 6th. Look at that precision in the second chart.

Image

Image
It's been fun for sure. Wild market moves and my accounts hold steady overall (bond heavy accounts are up). I've been taking the opportunity to heavily rebalance the cash proceeds from treasury futures into equities.
comeinvest
Posts: 2669
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest »

skierincolorado wrote: Thu Mar 16, 2023 12:51 pm Did you see they will have a micro mid cap starting March 20th? I am curious what the implied rate will be. If I understand, emd is 250k per contract which I don't quite have enough for. I'm getting close but I'm also hesitant to switch 250k straight from MES to EMD all at once.
Good news. I'm surprised, because the /EMD itself is not that liquid. (The calendar rolls are efficient though, which matters to us investors.)
My guess is that the micro midcap will mimic the /EMD accurately and even the bid/ask spread will be similar, because traders and market makers will arbitrage the two. So good news.
I am pocketing ca. 0.4% p.a. from my large cap / mid cap index arbitrage pair trade based on the difference in implied rates. I certainly have some industry concentration that way namely underexposure to big tech, which will result in some fluctuations; but the risk of the pair trade should be random i.e. totally uncorrelated to any other markets, with 0.4% p.a. expected excess return. In my other account I hedge the mid cap / large cap divergence risk with a large cap / mid cap / small cap offset. Theoretically the short Russell 2000 offset position should add to my market-neutral overlay returns in the long run, if the rumor materializes that the Russell is trash because of structural issues.

You will probably need to ask IB to add the micro via a web ticket, if you want to have it fast.
Last edited by comeinvest on Thu Mar 16, 2023 2:42 pm, edited 2 times in total.
comeinvest
Posts: 2669
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest »

skierincolorado wrote: Thu Mar 16, 2023 12:53 pm Wow expensive. Why buy msci futures at all? I might buy just 1 mxef for diversification and dollar hedging.
What is the alternative? My asset allocation is about world market cap, for maximal diversification benefits.
I think the DJ600 futures have less implied financing, but more collateral drag.
For emerging markets, you would have a drag of about 0.4% p.a. from dividend withholding if you implement the exposure without futures, that you have to net with the futures implied financing cost. That is before U.S. tax; in taxable accounts the situation is more complicated.
Options box spread currently have ca 0.3% financing spread above T-bills. In IRAs you can't use options box spreads; it is not clear if options box spreads result in unrelated business taxable income in 401(k) accounts.

I look at the puzzle as a "portable alpha" optimization problem - implement via outright equities or ETFs whatever has the maximum alpha vs the index implemented with futures; implement the rest of the asset allocation with futures to achieve your targets.
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 »

comeinvest wrote: Thu Mar 16, 2023 1:28 pm
skierincolorado wrote: Thu Mar 16, 2023 12:53 pm Wow expensive. Why buy msci futures at all? I might buy just 1 mxef for diversification and dollar hedging.
What is the alternative? My asset allocation is about world market cap, for maximal diversification benefits.
I think the DJ600 futures have less implied financing, but more collateral drag.
For emerging markets, you would have a drag of about 0.4% p.a. from dividend withholding if you implement the exposure without futures, that you have to net with the futures implied financing cost. That is before U.S. tax; in taxable accounts the situation is more complicated.
Options box spread currently have ca 0.3% financing spread above T-bills. In IRAs you can't use options box spreads; it is not clear if options box spreads result in unrelated business taxable income in 401(k) accounts.

I look at the puzzle as a "portable alpha" optimization problem - implement via outright equities or ETFs whatever has the maximum alpha vs the index implemented with futures; implement the rest of the asset allocation with futures to achieve your targets.
With DJ600 you get drag of 2% on 10% of the future exposure? Isn't that only like .2%?

What is the .4% drag on dividends withholding from?

DJ600 and nik225 and then either do emerging with ETFs or a little msci futures if necessary
comeinvest
Posts: 2669
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest »

skierincolorado wrote: Thu Mar 16, 2023 1:58 pm
comeinvest wrote: Thu Mar 16, 2023 1:28 pm
skierincolorado wrote: Thu Mar 16, 2023 12:53 pm Wow expensive. Why buy msci futures at all? I might buy just 1 mxef for diversification and dollar hedging.
What is the alternative? My asset allocation is about world market cap, for maximal diversification benefits.
I think the DJ600 futures have less implied financing, but more collateral drag.
For emerging markets, you would have a drag of about 0.4% p.a. from dividend withholding if you implement the exposure without futures, that you have to net with the futures implied financing cost. That is before U.S. tax; in taxable accounts the situation is more complicated.
Options box spread currently have ca 0.3% financing spread above T-bills. In IRAs you can't use options box spreads; it is not clear if options box spreads result in unrelated business taxable income in 401(k) accounts.

I look at the puzzle as a "portable alpha" optimization problem - implement via outright equities or ETFs whatever has the maximum alpha vs the index implemented with futures; implement the rest of the asset allocation with futures to achieve your targets.
With DJ600 you get drag of 2% on 10% of the future exposure? Isn't that only like .2%?

What is the .4% drag on dividends withholding from?

DJ600 and nik225 and then either do emerging with ETFs or a little msci futures if necessary
Yes Nikkei225 futures seem to be very efficient, at least before JPY short-term rates rise - then you will have a cash collateral drag problem again. But that's not expected to happen for another few years, at least based on the current JPY yield curve.

EUR rates per yield curve were 3% for the next few years, now dropped to about 2.75% for the next year on average.

I think EM stocks have about 4% dividend yield, and >10% average withholding rates. I'm typing this from my head; check the details.

To optimize my cost, I tend to put developed market stocks in my 401k because from most countries I get a refund of the entire withholding tax per 0% treaty rates for pensions and 401(k) accounts. Some countries (U.K.) have no withholding. If your account is small you may say it's not worth the time; but once the account gets larger we're talking about serious money.
From EM you don't get refunds.
It's also hard to replicate the EM index via a representative sampling because most stocks are not accessible. Much easier for DM. I do it in my taxable account.
So in summary, for EM I currently use futures and some small cap value ETFs in the tax-advantaged accounts, and some EM ADRs in taxable. For DM I use futures in tax-advantaged, and individual stocks in both tax-advantaged and taxable.
Last edited by comeinvest on Thu Mar 16, 2023 2:31 pm, edited 5 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 »

comeinvest wrote: Thu Mar 16, 2023 2:05 pm
skierincolorado wrote: Thu Mar 16, 2023 1:58 pm
comeinvest wrote: Thu Mar 16, 2023 1:28 pm
skierincolorado wrote: Thu Mar 16, 2023 12:53 pm Wow expensive. Why buy msci futures at all? I might buy just 1 mxef for diversification and dollar hedging.
What is the alternative? My asset allocation is about world market cap, for maximal diversification benefits.
I think the DJ600 futures have less implied financing, but more collateral drag.
For emerging markets, you would have a drag of about 0.4% p.a. from dividend withholding if you implement the exposure without futures, that you have to net with the futures implied financing cost. That is before U.S. tax; in taxable accounts the situation is more complicated.
Options box spread currently have ca 0.3% financing spread above T-bills. In IRAs you can't use options box spreads; it is not clear if options box spreads result in unrelated business taxable income in 401(k) accounts.

I look at the puzzle as a "portable alpha" optimization problem - implement via outright equities or ETFs whatever has the maximum alpha vs the index implemented with futures; implement the rest of the asset allocation with futures to achieve your targets.
With DJ600 you get drag of 2% on 10% of the future exposure? Isn't that only like .2%?

What is the .4% drag on dividends withholding from?

DJ600 and nik225 and then either do emerging with ETFs or a little msci futures if necessary
Yes Nikkei225 futures seem to be very efficient, at least before JPY short-term rates rise - then you will have a collateral cash drag problem again. But that's not expected to happen for another few years, at least based on the current JPY yield curve.

EUR rates per yield curve were 3% for the next few years, now dropped to about 2.75% for the next year on average.

I think EM stocks have about 4% dividend yield, and >10% average withholding rates. I'm typing this from my head; check the details.

To optimize my cost, I tend to put developed market stocks in my 401k because from most countries I get a refund of the entire withholding tax per 0% treaty rates for pensions and 401(k) accounts. Some countries (U.K.) have no withholding. If your account is small you may say it's not worth the time; but once the account gets larger we're talking about serious money.
From EM you don't get refunds.
It's also hard to replicate the EM index via a representative sampling because most stocks are not accessible. Much easier for DM. I do it in my taxable account.
Essentially you hold EM in taxable to capture the higher foreign tax credit. Or use futures in tax advantaged.

Buy the developed market msci futures don't have much use because dj600 and nik225 are cheaper.
comeinvest
Posts: 2669
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest »

skierincolorado wrote: Thu Mar 16, 2023 2:22 pm Essentially you hold EM in taxable to capture the higher foreign tax credit. Or use futures in tax advantaged.
Yes that's part of my strategy, but EM is hard to replicate and I don't want to use ETFs in taxable, that's my principle. I don't want to be locked into a fund once it has unrealized gains, and funds can also be an international tax headache if you ever move abroad. I added some thoughts about my tax strategy to my previous post after you cited it.
comeinvest
Posts: 2669
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest »

skierincolorado wrote: Thu Mar 16, 2023 2:22 pm Essentially you hold EM in taxable to capture the higher foreign tax credit. Or use futures in tax advantaged.

Buy the developed market msci futures don't have much use because dj600 and nik225 are cheaper.
Correct, but DJ600 has a cash collateral drag, unless IB lets us know if any government bonds can be used as futures collateral in lieu of cash. I currently use a mix of DJ600+Nikkei225 and MXEA, until I figure out which is cheaper in the long run.
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 »

comeinvest wrote: Thu Mar 16, 2023 2:34 pm
skierincolorado wrote: Thu Mar 16, 2023 2:22 pm Essentially you hold EM in taxable to capture the higher foreign tax credit. Or use futures in tax advantaged.

Buy the developed market msci futures don't have much use because dj600 and nik225 are cheaper.
Correct, but DJ600 has a cash collateral drag, unless IB lets us know if any government bonds can be used as futures collateral in lieu of cash. I currently use a mix of DJ600+Nikkei225 and MXEA, until I figure out which is cheaper in the long run.
Yeah I suppose the cash drag makes it closer but the rates on mxea are pretty clearly higher if what you posted is correct
comeinvest
Posts: 2669
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest »

skierincolorado wrote: Thu Mar 16, 2023 2:39 pm
comeinvest wrote: Thu Mar 16, 2023 2:34 pm
skierincolorado wrote: Thu Mar 16, 2023 2:22 pm Essentially you hold EM in taxable to capture the higher foreign tax credit. Or use futures in tax advantaged.

Buy the developed market msci futures don't have much use because dj600 and nik225 are cheaper.
Correct, but DJ600 has a cash collateral drag, unless IB lets us know if any government bonds can be used as futures collateral in lieu of cash. I currently use a mix of DJ600+Nikkei225 and MXEA, until I figure out which is cheaper in the long run.
Yeah I suppose the cash drag makes it closer but the rates on mxea are pretty clearly higher if what you posted is correct
The implied financing of MXEA and MXEF fluctuates every quarter. You can go to the ICE web site, they have the roll data more than 10 years back. But on average it's higher than other futures. I'll do some statistical analysis when I find some time.
I couldn't find historical or current EUREX roll data except in some older summary papers. Let me know if you can locate roll information for EUREX equity index futures.
But I went through the trouble and did some estimations based on comparisons with ETFs if you go a few pages back in this thread.
comeinvest
Posts: 2669
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest »

skierincolorado wrote: Thu Mar 16, 2023 12:51 pm
comeinvest wrote: Wed Mar 15, 2023 3:14 am On 03/13 the difference of S&P 400 and S&P 500 implied rate widened again to a relatively typical 0.41%.
It looks like I lucked out and got and additional 8.5 bps with my fills, locking in a 0.495% p.a. spread difference with my offsetting equity futures pair for the next quarter, if my math is right.

Image

Image
Did you see they will have a micro mid cap starting March 20th? I am curious what the implied rate will be. If I understand, emd is 250k per contract which I don't quite have enough for. I'm getting close but I'm also hesitant to switch 250k straight from MES to EMD all at once.
There will also be a micro S&P 600 small cap futures with the same launch date. The /SMC is very illiquid, so I'm not sure what is the point of the micro. I haven't looked at the calendar roll bid/ask spreads.

I don't know what index the midcap and smallcap asset classes in PV refer to. For whatever reason, the ETF IVOO shows a different max drawdown than the midcap asset class for the 2011+ period.

Interesting paper: https://www.spglobal.com/spdji/en/docum ... ations.pdf
Blog: https://www.indexologyblog.com/2022/06/ ... ooked-gem/

The performance and risk statistics are probably not extremely meaningful, because most of the divergence of performance happened at rather singular events - tech bubble, GFC, Covid, etc. But mid caps and small caps have definitely higher market beta, so when using midcap futures this needs to be accounted for in the asset allocation.
Also read this thread: viewtopic.php?t=281359

The last chart is of a period when small cap value (VBR) and market / growth (VB) had about equal performance.

Image

Image

Image

Image

Added the following charts on 03/19/2023:

Image

Image

Image

Image

Image

Image

Image

Added on 04/30/2023:

Image

Image
Last edited by comeinvest on Sun Apr 30, 2023 7:38 am, edited 4 times in total.
User avatar
cos
Posts: 506
Joined: Fri Aug 23, 2019 7:34 pm
Location: Boston
Contact:

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

Post by cos »

comeinvest wrote: Fri Mar 17, 2023 5:20 am I don't know what index the midcap and smallcap asset classes in PV refer to.
Portfolio Visualizer wrote:
US Mid Cap
Professor Kenneth French's Research Data 1972-1998
Vanguard Mid Cap Index Fund (VIMSX) 1999+

US Small Cap
Professor Kenneth French's Research Data 1972-1989
Vanguard Small Cap Index Fund (NAESX) 1990+
comeinvest wrote: Fri Mar 17, 2023 5:20 am For whatever reason, the ETF IVOO shows a different max drawdown than the midcap asset class for the 2011+ period.
This makes sense given IVOO's significantly greater size, value, and profitability loadings relative to VIMSX (thanks to the S&P's quality screen).
km91
Posts: 1373
Joined: Wed Oct 13, 2021 12:32 pm

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

Post by km91 »

comeinvest wrote: Fri Mar 17, 2023 5:20 am There will also be a micro S&P 600 small cap futures with the same launch date. The /SMC is very illiquid, so I'm not sure what is the point of the micro. I haven't looked at the calendar roll bid/ask spreads.
Why bother venturing into the more thinly traded indexes, instead of just holding a fund directly? Treasuries and ES futures should be enough leverage to be able to get the more specific exposures in the portfolio unlevered. You can get 4 or 5x leverage in an IRA with 75% of NAV in a SCV fund
comeinvest
Posts: 2669
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest »

km91 wrote: Sat Mar 18, 2023 12:08 am
comeinvest wrote: Fri Mar 17, 2023 5:20 am There will also be a micro S&P 600 small cap futures with the same launch date. The /SMC is very illiquid, so I'm not sure what is the point of the micro. I haven't looked at the calendar roll bid/ask spreads.
Why bother venturing into the more thinly traded indexes, instead of just holding a fund directly? Treasuries and ES futures should be enough leverage to be able to get the more specific exposures in the portfolio unlevered. You can get 4 or 5x leverage in an IRA with 75% of NAV in a SCV fund
Yes 75% of NAV in off-index positions and tilts, and 25% in collateral would work in IB IRA's with 2x margin requirement; although it would require relatively frequent monitoring and rebalancing. In trust accounts it's a bit better.

Because of the high financing cost of /ES (0.4-0.5% above risk-free rates last time I did a longer-term analysis), I go a different route. I use treasury and SOFR futures, /EMD (soon I will use the micro midcap futures that skier mentioned), AVUS, AVUV, and VBR in my tax-advantaged accounts. /EMD has negligible roll cost in terms of bid/ask spread, and implied financing at about T-bill rates for as long as I have monitored the contract. I also have international equities with 0% withholding tax rates in my tax-advantaged mHFEA accounts, along with some EM SCV ETFs (AVES, DGS, EYLD). I think this gives me the best diversification and the best spread of "portable alpha" in terms of implied rates and tax drag vs the equivalent index futures of each equity asset subclass. I learned the concept of portable alpha from public pensions and endowments.

The different financing rates of the different asset classes are a bit hard to explain; but we don't have to explain them, they are a fact of life. They provide a risk-free lunch, when shuffling around the implementation of the asset allocation accordingly.
comeinvest
Posts: 2669
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest »

I'm curious about the risk and return of ITT vs LTT depending on the initial slope of the yield curve.
I am comparing 150% equities / 250% ITT vs. 150% equities / 82% LTT, using VFITX for ITT and VUSTX for LTT. The 250% ITT and 82% LTT have equal duration based on current duration exposure (5.3 years for VFITX and 16.2 years for VUSTX). Unfortunately I can't locate historical durations, so the ratio might have been different in 2020 and in 2022.

Treasuries upside performance during a macroeconomic crash starting with a positively sloped yield curve:
On 02/28/2020, 5-year treasuries yielded 0.89%, and 30-year treasuries yielded 1.65%.
The performance in Mar 2020 was ever so slightly in favor of ITT - almost a tie.

Treasuries upside performance during a small macroeconomic mini-crash starting with a flat to slightly negatively sloped yield curve:
Free PV results not yet available for month-to-date.
On Feb 21 2023, 5-year treasuries yielded 4.16%, and 30-year treasuries yielded 3.98%.
Between Feb 21 2023 and Mar 17 2023 the portfolio performance was significantly in favor of ITT:
150% VFINX + 250% VFITX: -1.88 * 1.50 + 3.04 * 2.50 -> 4.78%
150% VFINX + 82% VUSTX: -1.88 * 1.50 + 5.9 * 0.82 -> 2.02%

Treasuries downside risk in a rising rates environment starting with a positively sloped yield curve:
Between 12/31/2021 and 12/30/2022, 5-year rates rose from 1.26% to 3.99%; 30-year rates rose from 1.9% to 3.97%.
The portfolio performance was nearly identical.

Summary: When the yield curve is flattish, short maturity treasuries might give the biggest upside bang for the risk.

If I had historical duration data, I could do much better modeling and backtesting.

Image

Image

Image
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 »

km91 wrote: Sat Mar 18, 2023 12:08 am
comeinvest wrote: Fri Mar 17, 2023 5:20 am There will also be a micro S&P 600 small cap futures with the same launch date. The /SMC is very illiquid, so I'm not sure what is the point of the micro. I haven't looked at the calendar roll bid/ask spreads.
Why bother venturing into the more thinly traded indexes, instead of just holding a fund directly? Treasuries and ES futures should be enough leverage to be able to get the more specific exposures in the portfolio unlevered. You can get 4 or 5x leverage in an IRA with 75% of NAV in a SCV fund
Only if you were 100% U.S.

I also do international at 35% of assets, which when leveraged 1.5x becomes 52.5%. Then factor tilts. Then treasury futures. Then consider it's an IRA and the futures collateral requirement is doubled at IB. If you use foreign denominated futures to get foreign equity exposure, then you need to short the dollar to get the same performance as VXUS which is dollar denominated. And then consider that if markets crashed I want the flexibility to increase my leverage to 2x per lifecycle investing principles. It doesn't all fit without using mid-cap futures, foreign futures, and currency futures to short the dollar.

Or you end up overweight in large cap and/or sacrificing factor tilts.
comeinvest
Posts: 2669
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest »

skierincolorado wrote: Sat Mar 18, 2023 8:01 pm
km91 wrote: Sat Mar 18, 2023 12:08 am
comeinvest wrote: Fri Mar 17, 2023 5:20 am There will also be a micro S&P 600 small cap futures with the same launch date. The /SMC is very illiquid, so I'm not sure what is the point of the micro. I haven't looked at the calendar roll bid/ask spreads.
Why bother venturing into the more thinly traded indexes, instead of just holding a fund directly? Treasuries and ES futures should be enough leverage to be able to get the more specific exposures in the portfolio unlevered. You can get 4 or 5x leverage in an IRA with 75% of NAV in a SCV fund
Only if you were 100% U.S.

I also do international at 35% of assets, which when leveraged 1.5x becomes 52.5%. Then factor tilts. Then treasury futures. Then consider it's an IRA and the futures collateral requirement is doubled at IB. If you use foreign denominated futures to get foreign equity exposure, then you need to short the dollar to get the same performance as VXUS which is dollar denominated. And then consider that if markets crashed I want the flexibility to increase my leverage to 2x per lifecycle investing principles. It doesn't all fit without using mid-cap futures, foreign futures, and currency futures to short the dollar.

Or you end up overweight in large cap and/or sacrificing factor tilts.
That's all correct. Like I said, logic dictates implementing without futures whatever has the highest expected alpha (mostly from implied financing cost and withholding taxes) in relation to the respective futures, in decreasing order, until the available cash minus the required futures collateral is consumed. The rest is implemented with futures to achieve the target asset allocation.

The future alpha from factor tilts of course is not known, which makes the puzzle a bit more challenging. But a pretty nicely diversified asset allocation can be achieved with the futures contracts available in the market, and spot positions with the available cash.
comeinvest
Posts: 2669
Joined: Mon Mar 12, 2012 6:57 pm

Trend following

Post by comeinvest »

Should I add trend following to my mHFEA?
It helped big time in 3 of the 4 major stock and bond market drawdowns since 2000, and was neutral in one. Based on this, it should be a great addition to mHFEA, shouldn't it?

My understanding is that the more drawn-out a downturn, the more effective trend following will be at hedging the effect on the portfolio. For the tail risk where it would amplify a sudden large catastrophic event, see the 1987 crash in the chart further down. The realized risk was not large.

Charts from the paper: https://www.aqr.com/Insights/Trend-Following

Image

Image

Benefits to a portfolio from the paper https://www.grahamcapital.com/assets/In ... 202022.pdf

Image

Image

From the paper https://www.aqr.com/Insights/Research/J ... -Investing :

Image

Image

Image

Image

From KraneShares.com:

Image

Update 2023-04-15: Charts from InvestReSolve Twitter account https://twitter.com/InvestReSolve/statu ... 7969176576 :

Image

Image

mHFEA performance with vs without an overlay of trend following: https://www.portfoliovisualizer.com/bac ... tion5_2=10
The addition of 10% DBMF, 10% KMLM reduced the 2022 drawdown from -56% to -52%.
Last edited by comeinvest on Sat Apr 15, 2023 9:46 pm, edited 5 times in total.
comeinvest
Posts: 2669
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest »

skierincolorado wrote: Sat Mar 18, 2023 8:01 pm
km91 wrote: Sat Mar 18, 2023 12:08 am
comeinvest wrote: Fri Mar 17, 2023 5:20 am There will also be a micro S&P 600 small cap futures with the same launch date. The /SMC is very illiquid, so I'm not sure what is the point of the micro. I haven't looked at the calendar roll bid/ask spreads.
Why bother venturing into the more thinly traded indexes, instead of just holding a fund directly? Treasuries and ES futures should be enough leverage to be able to get the more specific exposures in the portfolio unlevered. You can get 4 or 5x leverage in an IRA with 75% of NAV in a SCV fund
Only if you were 100% U.S.

I also do international at 35% of assets, which when leveraged 1.5x becomes 52.5%. Then factor tilts. Then treasury futures. Then consider it's an IRA and the futures collateral requirement is doubled at IB. If you use foreign denominated futures to get foreign equity exposure, then you need to short the dollar to get the same performance as VXUS which is dollar denominated. And then consider that if markets crashed I want the flexibility to increase my leverage to 2x per lifecycle investing principles. It doesn't all fit without using mid-cap futures, foreign futures, and currency futures to short the dollar.

Or you end up overweight in large cap and/or sacrificing factor tilts.
Don't un-hedge the entire foreign currency exposure from foreign equity futures. That might be ok in the long run for non-leveraged investors as currencies are thought to be a zero sum game in the long run; but with a leveraged strategy I think you might have less volatility and drawdowns with partial hedging.
comeinvest
Posts: 2669
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest »

Why did the June maturity T-bill charts not reflect the Mar 2023 SOFR movement between 03/15 and 03/16 ? The Mar 2023 SOFR expires on 06/21.

Image

Image

Image

Image
impatientInv
Posts: 349
Joined: Fri Sep 03, 2021 1:26 pm

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

Post by impatientInv »

comeinvest wrote: Thu Mar 09, 2023 10:54 pm On some days, the mHFEA portfolio actually behaves as designed. So much fun.

Image
I see you are using the EMD - E-mini S&P MidCap 400 futures. Hoping that MMC - micro E-mini S&P MidCap 400 futures is available on IBKR soon, it was launched today, Hope it will have enough volume and is popular.

Don't see any volume in SMC - E-mini S&P SmallCap 600 futures. Is this correct? Have you bought this?

https://www.cmegroup.com/markets/equiti ... olume.html
No individual stocks.
comeinvest
Posts: 2669
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest »

impatientInv wrote: Mon Mar 20, 2023 9:01 pm I see you are using the EMD - E-mini S&P MidCap 400 futures. Hoping that MMC - micro E-mini S&P MidCap 400 futures is available on IBKR soon, it was launched today, Hope it will have enough volume and is popular.

Don't see any volume in SMC - E-mini S&P SmallCap 600 futures. Is this correct? Have you bought this?

https://www.cmegroup.com/markets/equiti ... olume.html
No, I'm not sure why they are adding a micro smallcap when there is no volume for the mini smallcap.

If you create a product request ticket for the micro midcap, they might add it faster.
impatientInv
Posts: 349
Joined: Fri Sep 03, 2021 1:26 pm

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

Post by impatientInv »

comeinvest wrote: Mon Mar 20, 2023 11:05 pm
impatientInv wrote: Mon Mar 20, 2023 9:01 pm I see you are using the EMD - E-mini S&P MidCap 400 futures. Hoping that MMC - micro E-mini S&P MidCap 400 futures is available on IBKR soon, it was launched today, Hope it will have enough volume and is popular.

Don't see any volume in SMC - E-mini S&P SmallCap 600 futures. Is this correct? Have you bought this?

https://www.cmegroup.com/markets/equiti ... olume.html
No, I'm not sure why they are adding a micro smallcap when there is no volume for the mini smallcap.

If you create a product request ticket for the micro midcap, they might add it faster.
The volume on Micro E-mini S&P MidCap 400 futures is quite low. Volume was 40 on Mar 20th and zero of 21st.

Is there a problem with buying such a low volume product? Anyone have experience with this kind of futures? Would a limit orders be good enough?

https://www.cmegroup.com/markets/equiti ... olume.html
No individual stocks.
comeinvest
Posts: 2669
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest »

impatientInv wrote: Wed Mar 22, 2023 6:25 pm
comeinvest wrote: Mon Mar 20, 2023 11:05 pm
impatientInv wrote: Mon Mar 20, 2023 9:01 pm I see you are using the EMD - E-mini S&P MidCap 400 futures. Hoping that MMC - micro E-mini S&P MidCap 400 futures is available on IBKR soon, it was launched today, Hope it will have enough volume and is popular.

Don't see any volume in SMC - E-mini S&P SmallCap 600 futures. Is this correct? Have you bought this?

https://www.cmegroup.com/markets/equiti ... olume.html
No, I'm not sure why they are adding a micro smallcap when there is no volume for the mini smallcap.

If you create a product request ticket for the micro midcap, they might add it faster.
The volume on Micro E-mini S&P MidCap 400 futures is quite low. Volume was 40 on Mar 20th and zero of 21st.

Is there a problem with buying such a low volume product? Anyone have experience with this kind of futures? Would a limit orders be good enough?

https://www.cmegroup.com/markets/equiti ... olume.html
Not necessarily a problem. You have to monitor the real time bid/ask spread during market hours. I suppose they might be similar to that of /EMD, because dealers will arbitrage the two so they will be tied to each other.
Then you would have to look at the calendar roll spread come June, which matters more than the outright spread if you buy and hold (buy and roll). I would also think that the micro's calendar spread bid/ask is almost as tight as that of the mini. The mini calendar spread is at the minimum tick size during the days with highest roll volume, if I remember right.
comeinvest
Posts: 2669
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest »

Perhaps an alternative to complicated leveraged and rebalanced multi-asset strategies would be to put everything into some solid higher beta index, then wait until retirement and chill. No worries about cost of leverage, margin calls, deleveraging, and what not.

Mid-cap for example would have resulted in twice the terminal balance of the S&P 500. Could complement this with some small-cap value ETF like AVUV perhaps.

Image
impatientInv
Posts: 349
Joined: Fri Sep 03, 2021 1:26 pm

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

Post by impatientInv »

comeinvest wrote: Mon Mar 20, 2023 11:05 pm
impatientInv wrote: Mon Mar 20, 2023 9:01 pm I see you are using the EMD - E-mini S&P MidCap 400 futures. Hoping that MMC - micro E-mini S&P MidCap 400 futures is available on IBKR soon, it was launched today, Hope it will have enough volume and is popular.

Don't see any volume in SMC - E-mini S&P SmallCap 600 futures. Is this correct? Have you bought this?

https://www.cmegroup.com/markets/equiti ... olume.html
No, I'm not sure why they are adding a micro smallcap when there is no volume for the mini smallcap.

If you create a product request ticket for the micro midcap, they might add it faster.
They still have not created the Micro E-mini S&P MidCap 400 future on IBKR. the person answering ticket was confused between mini and micro..
No individual stocks.
comeinvest
Posts: 2669
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest »

impatientInv wrote: Sat Mar 25, 2023 1:33 pm They still have not created the Micro E-mini S&P MidCap 400 future on IBKR. the person answering ticket was confused between mini and micro..
Be persistent and impatient. If we all put in requests, perhaps they have it by June.
director84
Posts: 49
Joined: Thu Dec 23, 2010 9:59 am

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

Post by director84 »

First of all, I'd like to say a big thank you to skierincolorado, MillenialMillions, Siamond, and others who have contributed so much to this and similar threads. I haven't met you, but I've learned an enormous amount from reading your posts over the past months and years. It's very much appreciated how much time and brainpower you've shared with this community.

I'm bought into this strategy but have a few implementation specific questions that I haven't been able to find answers to.

1. Regarding box spreads, what execution date do you target? Do you just look for the one that provides the lowest rate (which is currently Dec '27 according to boxtrades.com). If rates were to drop before then, can you "refinance" by closing the box early and selling a new one?

2. How do you manage your margin cushion with high leverage? Do you rebalance leverage periodically (e.g. quarterly) AND when you get within say 10% of maintenance margin? Specifically, I'm considering 125/200 stock/ITT allocation but have seen skierincolorado mention even higher targets. With 325% leverage and a 25% maintenance* requirement, the portfolio can only drop ~19% before a margin call. (In reality, I have a 401k that doesn't allow trading options or leveraged ETFs so my leverage in the other accounts has to be even higher to make up for it, further reducing my cushion.)
(*I don't know exactly what the maintenance requirement is on portfolio margin. I've seen 25% referenced a few times, but perhaps it's lower or higher.)
comeinvest
Posts: 2669
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest »

director84 wrote: Sat Mar 25, 2023 2:48 pm First of all, I'd like to say a big thank you to skierincolorado, MillenialMillions, Siamond, and others who have contributed so much to this and similar threads. I haven't met you, but I've learned an enormous amount from reading your posts over the past months and years. It's very much appreciated how much time and brainpower you've shared with this community.

I'm bought into this strategy but have a few implementation specific questions that I haven't been able to find answers to.

1. Regarding box spreads, what execution date do you target? Do you just look for the one that provides the lowest rate (which is currently Dec '27 according to boxtrades.com). If rates were to drop before then, can you "refinance" by closing the box early and selling a new one?
No. For purpose of your implementation of leverage, ignore the yield curve. Don't try to interpret it, or to micro-manage the timing or the maturities of your trades. The yield curve is mostly explained by the expectations hypothesis, which would dictate that the maturity of your box doesn't matter. I think most of us go with a 3-month cycle, using the 3rd Friday or end-of-month expirations. Going with somewhat longer expiration would be ok, but watch as the margin increases for longer expirations. For 3 months, the margin is close to zero. Then you also have the flexibility to change your exposure every 3 months, without incurring additional cost.
I am mostly converting to bear puts with high strike prices, instead of box spreads. Bear puts have only 2 put legs, omitting the 2 call legs of the box. For strike prices above 6000 I get at least the same rates that I get with boxes.
director84 wrote: Sat Mar 25, 2023 2:48 pm 2. How do you manage your margin cushion with high leverage? Do you rebalance leverage periodically (e.g. quarterly) AND when you get within say 10% of maintenance margin? Specifically, I'm considering 125/200 stock/ITT allocation but have seen skierincolorado mention even higher targets. With 325% leverage and a 25% maintenance* requirement, the portfolio can only drop ~19% before a margin call. (In reality, I have a 401k that doesn't allow trading options or leveraged ETFs so my leverage in the other accounts has to be even higher to make up for it, further reducing my cushion.)
(*I don't know exactly what the maintenance requirement is on portfolio margin. I've seen 25% referenced a few times, but perhaps it's lower or higher.)
Regarding the amount of leverage, there are many posts about this, and no one right answer. Also read the leveraged lifecycle investing thread.

Regarding rebalancing strategies, I think there is no consensus. Also read the lifecycle thread, where we discussed this on a few pages worth of posts. My takeaway so far is that, ignoring lifecycle age and asset based target allocations for the moment, the volatility decay is mostly controlled by the leverage ratio, and not by the frequency of rebalancing. The frequency of rebalancing is mostly a risk/return tradeoff: More frequent rebalancing is slightly less risky with slightly lower returns. There are other threads that showed that for any rebalancing frequency > 3 months, the results are about the same and converge over long time periods; but more frequent rebalancing (e.g. 3 months vs. 12 months) has more consistent results. I think this was shown for both unleveraged stock/bond portfolios, as well as for HFEA if you read the HFEA thread.

With portfolio margin, most stocks and ETFs have about 15% maintenance margin; equity index futures have about 5-10%; treasury futures about 3% depending on maturity of the underlying. What that means is that those numbers are irrelevant (unless you run unrelated overlay strategies that consume additional margin), because you want to deleverage long time before you reach those margin call limits, to avoid volatility decay.

The margin requirements can go up, like double or triple, during equity market drawdowns. So plan ahead.

Regarding leveraging parts of the portfolio with higher leverage ratios to compensate for non-leverageable assets: We discussed this earlier in this thread at length. In short words, the non-leverageable assets won't help with volatility decay of the leverageable bucket, in case you need or want to deleverage at some point during a drawdown to reduce risk; so if vol decay is the constraint, I would ignore the non-leverageable assets. For low overall leverage ratios, and/or if the money is somewhat fungible between the buckets, and/or if the time horizon is shorter, the expected terminal total value of all assets is more relevant than the individual buckets, and the answer is somewhere in between this and basing the leverage on the overall amount of assets only. But again if you read the entire thread, you will find a longer discussion regarding this with examples.
comeinvest
Posts: 2669
Joined: Mon Mar 12, 2012 6:57 pm

Commodity futures

Post by comeinvest »

I would like to link this post about a book from Antti Ilmanen that is, among other things, about independent sources of returns, somewhat related to mHFEA: viewtopic.php?t=400761

I have to look into what he says about commodity futures, if it could be a potential addition to mHFEA. For those of us who already use treasury, equity index, and SOFR futures, it should be easy to add a permanent rolling allocation to a diversified mix of commodity futures at near zero extra cost (zero expense ratio), if it were to improve risk-adjusted returns. (I'm not talking about trend-following here, which is certainly more labor intensive.)

EDIT 03/31/2023: The Credit Suisse Global Investment Returns Yearbook 2023 https://www.credit-suisse.com/about-us- ... 02302.html also says: "Our finding is that while investing in individual commodities have themselves yielded very low long returns, thanks to the power of diversification, portfolios of futures have provided attractive risk-adjusted long-term returns."
The studies show a long run commodity premium over cash of 3-5%.
Diversification is very important in the commodity space so don’t buy single commodities.
The volatility of any single commodity can be 30% or higher. The volatility drag can bring the geometric return of the commodity to zero over longer times.
Yet, a diversified portfolio of these commodities can earn 3% compound average return because of the lower volatility. Rebalancing Bonus.
Outlook for Commodity Futures: 3% over cash for a diversified commodity portfolio
Broad commodity portfolios are useful because they can hedge different types of inflation
Chart from the Credit Suisse Global Investment Returns Yearbook 2023 https://www.credit-suisse.com/about-us- ... 02302.html
Info: 12 ^ (1/122) -> ca. 1.0206

Image

Related paper: https://investresolve.com/wp-content/up ... dities.pdf

Also from the Credit Suisse 2023 Yearbook:

Image
Last edited by comeinvest on Fri Jun 02, 2023 9:46 pm, edited 5 times in total.
director84
Posts: 49
Joined: Thu Dec 23, 2010 9:59 am

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

Post by director84 »

Great response. Thanks for the detailed information. It's certainly answered my questions and given me a lot more to consider/research.
I think most of us go with a 3-month cycle, using the 3rd Friday or end-of-month expirations. Going with somewhat longer expiration would be ok, but watch as the margin increases for longer expirations. For 3 months, the margin is close to zero. Then you also have the flexibility to change your exposure every 3 months, without incurring additional cost.
I am mostly converting to bear puts with high strike prices, instead of box spreads. Bear puts have only 2 put legs, omitting the 2 call legs of the box. For strike prices above 6000 I get at least the same rates that I get with boxes.
Great insight, thanks. Just when I thought I was beginning to understand box spreads, the rabbit hole goes deeper... I found your related comments on the SPX Box Spreads thread and will digest that some more.
Regarding rebalancing strategies, I think there is no consensus. Also read the lifecycle thread, where we discussed this on a few pages worth of posts. My takeaway so far is that, ignoring lifecycle age and asset based target allocations for the moment, the volatility decay is mostly controlled by the leverage ratio, and not by the frequency of rebalancing. The frequency of rebalancing is mostly a risk/return tradeoff: More frequent rebalancing is slightly less risky with slightly lower returns. There are other threads that showed that for any rebalancing frequency > 3 months, the results are about the same and converge over long time periods; but more frequent rebalancing (e.g. 3 months vs. 12 months) has more consistent results. I think this was shown for both unleveraged stock/bond portfolios, as well as for HFEA if you read the HFEA thread.
I already have some of my portfolio invested in HFEA and am aware that quarterly rebalancing backtested the best. However, the leverage is automatically rebalanced daily, due to the nature of those ETFs, so it's not something I gave much thought to. I liked HFEA for it's simplicity, but now that I've read through more of this thread and have a better understanding of futures and options, I'm convinced it's the superior and still manageable approach, just took me a little more studying (and a bear market) to understand that :D
Regarding leveraging parts of the portfolio with higher leverage ratios to compensate for non-leverageable assets: We discussed this earlier in this thread at length. In short words, the non-leverageable assets won't help with volatility decay of the leverageable bucket, in case you need or want to deleverage at some point during a drawdown to reduce risk; so if vol decay is the constraint, I would ignore the non-leverageable assets. For low overall leverage ratios, and/or if the money is somewhat fungible between the buckets, and/or if the time horizon is shorter, the expected terminal total value of all assets is more relevant than the individual buckets, and the answer is somewhere in between this and basing the leverage on the overall amount of assets only. But again if you read the entire thread, you will find a longer discussion regarding this with examples.
Thanks for the summary. I'm planning to read the entire thread but haven't gotten through all of it, yet. I'll make sure not to miss this part.
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 »

director84 wrote: Sat Mar 25, 2023 2:48 pm First of all, I'd like to say a big thank you to skierincolorado, MillenialMillions, Siamond, and others who have contributed so much to this and similar threads. I haven't met you, but I've learned an enormous amount from reading your posts over the past months and years. It's very much appreciated how much time and brainpower you've shared with this community.

I'm bought into this strategy but have a few implementation specific questions that I haven't been able to find answers to.

1. Regarding box spreads, what execution date do you target? Do you just look for the one that provides the lowest rate (which is currently Dec '27 according to boxtrades.com). If rates were to drop before then, can you "refinance" by closing the box early and selling a new one?

2. How do you manage your margin cushion with high leverage? Do you rebalance leverage periodically (e.g. quarterly) AND when you get within say 10% of maintenance margin? Specifically, I'm considering 125/200 stock/ITT allocation but have seen skierincolorado mention even higher targets. With 325% leverage and a 25% maintenance* requirement, the portfolio can only drop ~19% before a margin call. (In reality, I have a 401k that doesn't allow trading options or leveraged ETFs so my leverage in the other accounts has to be even higher to make up for it, further reducing my cushion.)
(*I don't know exactly what the maintenance requirement is on portfolio margin. I've seen 25% referenced a few times, but perhaps it's lower or higher.)
In addition to comeinvests points, all of which are important, I'd also stress the principles of lifecycle investing. Practically this means when your wealth decreases, your leverage should go up according to a pre-defined formula.

I'd also point out that the 125/200 equity/ITT is more bond heavy than what I and several others eventually decided on later in the thread. I should probably edit some of the earlier posts to reflect that. Initially I was too focused on replicating the performance of regular HFEA in recent decades. Millenialmillions spreadsheet was helpful in this regard as well as Simbas. Personally I am currently around 150/180 where the 180 is calculated by adjusting all the bonds to the same duration as ZF (4.3 years - not 5 years). And the actual duration is a bit shorter than that, more like 4 years (mix of mostly ZF, 0-2yr SOFRs, and ZN).

I'd also put emphasis on comeinvest's point about not including non-leveragable assets into the calculation. Your plan to leverage overall to 1.25x is low enough that you might be able to partially include it in the calculation, depending on the relative account sizes. The limiting factor is basically would you get margin called in the leveraged account if the market crashed 50%+ trying to maintain your overall dynamic target leverage (which should actually increase in a crash according to lifecycle principles). Even with 1.25x overall leverage with accounts of equal size, there is risk in a large crash. In a 50% crash, your overall leverage should increase from 1.25x to ~1.5x, which could be tough to maintain only leveraging one account especially when that account is losing more money and shrinking in size relative to the unleveraged account.

I suggest coming up with some target AAs as a function of current wealth, and then running through some examples of whether you could maintain those targets under a 50%+ market drop.

For example, at 400k of wealth = 1.25x leverage = 500k invested (100k borrowed)
market drops 50%, current wealth is now 150-160k, current equity leverage target would be 1.5-1.67x per lifecycle principles (define the plan in advance. With no rebalancing it would be 1.67x (100k borrowed on 150k equity). Personally I don't plan to rebalance/sell much at all unless leverage gets high)

Practically speaking, could you achieve these targets? After simulating the above with per account detail (the losses in the leveraged account will be greater leaving less remaining to leverage)?

It also seems from watching behavior in the HFEA thread that it is tough for people psychologically to actually increase leverage when the market is down. People seem to do the opposite and bail on the strategy.
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 »

comeinvest wrote: Fri Mar 24, 2023 11:48 pm Perhaps an alternative to complicated leveraged and rebalanced multi-asset strategies would be to put everything into some solid higher beta index, then wait until retirement and chill. No worries about cost of leverage, margin calls, deleveraging, and what not.

Mid-cap for example would have resulted in twice the terminal balance of the S&P 500. Could complement this with some small-cap value ETF like AVUV perhaps.

Image
Beta is the enemy :twisted: haha

But in seriousness, the sharpe of the large cap and small cap is the same meaning you could leverage the large to get pretty close to the small total return. Mid may just be an anomaly. Would be interesting to test longer. My understanding is that over long horizons SCV has lower sharpe than the market. Small and probably Mid probably do too. Would also need to know how they are defining small/mid/large here.
comeinvest
Posts: 2669
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest »

skierincolorado wrote: Sat Mar 25, 2023 11:09 pm
comeinvest wrote: Fri Mar 24, 2023 11:48 pm Perhaps an alternative to complicated leveraged and rebalanced multi-asset strategies would be to put everything into some solid higher beta index, then wait until retirement and chill. No worries about cost of leverage, margin calls, deleveraging, and what not.

Mid-cap for example would have resulted in twice the terminal balance of the S&P 500. Could complement this with some small-cap value ETF like AVUV perhaps.

Image
Beta is the enemy :twisted: haha

But in seriousness, the sharpe of the large cap and small cap is the same meaning you could leverage the large to get pretty close to the small total return. Mid may just be an anomaly. Would be interesting to test longer. My understanding is that over long horizons SCV has lower sharpe than the market. Small and probably Mid probably do too. Would also need to know how they are defining small/mid/large here.
SCV has higher sharpe since 1975. From 1995-2018 (trying to exclude large cap tech anomalies) it's about the same. But "they" say that the factor returns are kind or uncorrelated to the market returns. I'm not posting more charts, because with every time frame you get different results. I would agree that I would not read too much into the higher sharpe of mid caps; it may be coincidence. But the mid caps have higher market beta, I just realized that when I monitored my mid cap futures. So it would make sense that in the long run the return will be higher, even if the sharpe is the same. What I was trying to say is, using higher beta indexes instead of leverage on the personal account level would eliminate a whole lot of trouble. You could just pour everything into an ETF and chill until retirement comes, wake up 30 years later never have to worry about margin or rolling anything, and enjoy possibly twice as much money as the S&P 500 folks, and who knows, possibly more than the mHFEA folks. Here is somebody who does something similar: viewtopic.php?t=372156 . I'm not going to do this, because I already know how to use futures at this point, and I probably want some diversification into large caps, and then there are also the treasuries.
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: Sat Mar 25, 2023 10:33 pm Practically this means when your wealth decreases, your leverage should go up according to a pre-defined formula.
Are you sure? I was under the impression that lifecycle investing means leverage at any given point in time is a function of your estimate of how much you’ll invest from that point till retirement, which correlates (inversely) with your current wealth but isn’t the same thing.
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 Mar 26, 2023 2:18 pm
skierincolorado wrote: Sat Mar 25, 2023 10:33 pm Practically this means when your wealth decreases, your leverage should go up according to a pre-defined formula.
Are you sure? I was under the impression that lifecycle investing means leverage at any given point in time is a function of your estimate of how much you’ll invest from that point till retirement, which correlates (inversely) with your current wealth but isn’t the same thing.
Yes I'm sure. Its a function of both, but how much you expect to make until retirement doesn't usually change much from month to month. Whereas your present net worth can change dramatically. The goal is to maintain a constant fraction of lifetime wealth (present and future wealth discounted to the present) invested in stocks. As an example, if I have 300k today and estimate a present value of future savings of 300k, then my combined present and future savings are 600k. If I wish to maintain 60% of that in stock, then I must invest 360k today. I must leverage my 300k to 360k (1.2x). If the market falls 50%, I will lose 180k and have 120k remaining. If the present value of my future expected earnings remain unchanged at 300k, then my combined present and future wealth is 120k(present) + 300k (present value of future savings) = 420k. I still want to maintain 60% of that in stocks = 252k. To do so I would leverage my remaining 120k 2.1x.

The authors recommend a cap on leverage of 2x (essentially this is because the higher you go the volatility decay becomes too extreme). You can also smooth things out so that the leverage doesn't increase quite so quickly in a drawdown. In the above example I probably wouldn't go all the way to 2.1x. Probably more like 1.7x. For very young investors with small investments (< 3 years savings), high savings rates, and secure jobs, going as high as 2.5x or 3x may improve results.

This concept is critical to understand for minimizing lifetime risk and maximizing median and mean outcomes.

This is the formula for leverage from the author's spreadsheet:

MIN(cap,(lambda1*O1197)*(@INDEX(Contributions!Contributions!$H$529,O$101)+O9)/(O9+@INDEX(Contributions!$G$2:$G$529,O$101)))

cap = 2x
lambda is the % allocation to stock in retirement aka Samuelson Share
O1197 is a shiller PE adjustment (optional)
Contributions!Contributions!$H$529,O$101)+O9 = present value of remaining contributions + current wealth
O9+@INDEX(Contributions!$G$2:$G$529,O$101 = present wealth + monthly contribution


Thus the formula (ignoring the Shiller PE adjustment) becomes:

(Samuelson Share * present value of remaining contributions + current wealth) / (present wealth)

Thus if the denominator, present wealth, decreases, there will be an increase in leverage. For example, if the numerator is 500k and the denominator is 400k, then leverage is 1.25x. If present wealth decreases to 300k, then leverage becomes 1.6x.
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 »

Just one quick example from the spreadsheet,

In one month they have 140.7k leveraged 1.38x to 194.1k (borrowing 54.4k)

The next month there is a 11.1% market loss

The next cell shows 119.4k leveraged 1.48x to 176.7k (borrowing 57.3k)


Constant leverage (1.38x) would require selling during a market drop. 119.4k leveraged 1.38x would be 164.8k (45.6k borrowed)

Constant borrowing requires doing nothing in a market drop. (119.4k, borrowing 54.4k to 173.8k, which would be 1.46x leverage)

The spreadsheet actually bought a little in the market drop. It's pretty close to constant borrowing, but it actually increased borrowing slightly, resulting in 1.48x leverage.

This is from a small 11% market move. Big market moves can easily take you from 1.2x to 2x+ leverage following lifecycle investing.

Constant fraction of present value of lifetime wealth in stocks.
director84
Posts: 49
Joined: Thu Dec 23, 2010 9:59 am

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

Post by director84 »

skierincolorado wrote: Sat Mar 25, 2023 10:33 pm
director84 wrote: Sat Mar 25, 2023 2:48 pm First of all, I'd like to say a big thank you to skierincolorado, MillenialMillions, Siamond, and others who have contributed so much to this and similar threads. I haven't met you, but I've learned an enormous amount from reading your posts over the past months and years. It's very much appreciated how much time and brainpower you've shared with this community.

I'm bought into this strategy but have a few implementation specific questions that I haven't been able to find answers to.

1. Regarding box spreads, what execution date do you target? Do you just look for the one that provides the lowest rate (which is currently Dec '27 according to boxtrades.com). If rates were to drop before then, can you "refinance" by closing the box early and selling a new one?

2. How do you manage your margin cushion with high leverage? Do you rebalance leverage periodically (e.g. quarterly) AND when you get within say 10% of maintenance margin? Specifically, I'm considering 125/200 stock/ITT allocation but have seen skierincolorado mention even higher targets. With 325% leverage and a 25% maintenance* requirement, the portfolio can only drop ~19% before a margin call. (In reality, I have a 401k that doesn't allow trading options or leveraged ETFs so my leverage in the other accounts has to be even higher to make up for it, further reducing my cushion.)
(*I don't know exactly what the maintenance requirement is on portfolio margin. I've seen 25% referenced a few times, but perhaps it's lower or higher.)
In addition to comeinvests points, all of which are important, I'd also stress the principles of lifecycle investing. Practically this means when your wealth decreases, your leverage should go up according to a pre-defined formula.

I'd also point out that the 125/200 equity/ITT is more bond heavy than what I and several others eventually decided on later in the thread. I should probably edit some of the earlier posts to reflect that. Initially I was too focused on replicating the performance of regular HFEA in recent decades. Millenialmillions spreadsheet was helpful in this regard as well as Simbas. Personally I am currently around 150/180 where the 180 is calculated by adjusting all the bonds to the same duration as ZF (4.3 years - not 5 years). And the actual duration is a bit shorter than that, more like 4 years (mix of mostly ZF, 0-2yr SOFRs, and ZN).

I'd also put emphasis on comeinvest's point about not including non-leveragable assets into the calculation. Your plan to leverage overall to 1.25x is low enough that you might be able to partially include it in the calculation, depending on the relative account sizes. The limiting factor is basically would you get margin called in the leveraged account if the market crashed 50%+ trying to maintain your overall dynamic target leverage (which should actually increase in a crash according to lifecycle principles). Even with 1.25x overall leverage with accounts of equal size, there is risk in a large crash. In a 50% crash, your overall leverage should increase from 1.25x to ~1.5x, which could be tough to maintain only leveraging one account especially when that account is losing more money and shrinking in size relative to the unleveraged account.

I suggest coming up with some target AAs as a function of current wealth, and then running through some examples of whether you could maintain those targets under a 50%+ market drop.

For example, at 400k of wealth = 1.25x leverage = 500k invested (100k borrowed)
market drops 50%, current wealth is now 150-160k, current equity leverage target would be 1.5-1.67x per lifecycle principles (define the plan in advance. With no rebalancing it would be 1.67x (100k borrowed on 150k equity). Personally I don't plan to rebalance/sell much at all unless leverage gets high)

Practically speaking, could you achieve these targets? After simulating the above with per account detail (the losses in the leveraged account will be greater leaving less remaining to leverage)?

It also seems from watching behavior in the HFEA thread that it is tough for people psychologically to actually increase leverage when the market is down. People seem to do the opposite and bail on the strategy.
Thanks skierincoloardo for the detailed response. You and comeinvest have certainly given me a lot to dig into. I've read through The SIMPLE Portfolio summary by MillenailMillions a few times, but I think there's a lot of nuance I'd still like to understand better if this is a strategy I plan to commit to for the next few decades. I spent a good chunk of the weekend reading through this thread and am still only about a third of the way in. There is a lot of great discussion (some still over my head) but it's been extremely educational. Two weeks ago, my knowledge of box spreads was approximately zero.

I'll probably come back here with more questions, and hopefully contributions, once I've caught up on everything.
Post Reply