Should I use margin to buy a balanced fund?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
long_gamma
Posts: 347
Joined: Mon Sep 14, 2015 4:13 pm

Re: Should I use margin to buy a balanced fund?

Post by long_gamma » Mon Dec 04, 2017 6:37 pm

Some what relevant article to this thread.

https://blog.thinknewfound.com/2017/12/ ... y-getting/
"Everyone has a plan 'till they get punched in the mouth." --Mike Tyson

Topic Author
Rob Bertram
Posts: 809
Joined: Mon May 05, 2014 12:15 pm

Re: Should I use margin to buy a balanced fund?

Post by Rob Bertram » Tue Dec 05, 2017 2:05 pm

Awesome read! It is very much in line with this thread.

I really do hope that leveraged portfolios becomes more popular. If one of the robo-advisers would offer leveraged portfolios as an investment option, I would definitely stop my manual stuff and move it over to their service. Rolling futures isn't time consuming, but it does require access to your account. That means I have to plan vacations around my portfolio which is mildly inconvenient.

Swelfie
Posts: 200
Joined: Mon Mar 14, 2016 12:54 am

Re: Should I use margin to buy a balanced fund?

Post by Swelfie » Tue Dec 05, 2017 3:18 pm

Having been running a similar risk-parity portfolio for over a year now, one thing I was a bit caught off guard by was that, as Rob had discussed earlier in this thread, rising interest rates aren't really all that troublesome to the short term bonds. But what I am seeing in this current market is that combined with a flattening yield curve they are. The last few months interest rates have risen at the 2 year point, but they have risen faster at the 3 month point. This has pushed the implied repo rate up at a rate that my 2 year treasuries have the same expected return as they did 4 months ago, but they have lost a lot of value. My original assessment was that if interest rates rise I would expect to increase future returns, but I hadn't really accounted for the effect curve slope could have as a risk unto itself.

Not that I'm dissatisfied. My portfolio is behaving well overall. I am wondering if there is a diversification path here though. Here is my thought:

VIX doesn't really have any expected future growth, rather it spikes and returns to a baseline. The "insurance" aspect of holding VIX directly against an equity crash causes prices on future volatility to be higher than expected. This puts VIX futures in contago almost always (except during market crashes) and conversely, puts short VIX in near permanent backwardation, making a sliver of ZIV, for instance an attractive investment.

Yield curves have a similar story to VIX in that normally the curve is somewhat steep. At times it shallows or inverts but it later returns to normal (I can't see an economy that persisted in a state where long term investments always pay more than short term.) Since straight-forward investment in a positively sloped yield curve, such as this thread illustrates, is the norm, then insurance against curve flattening is likely at a premium. What is the combination of derivatives then that is shorts a flattening yield curve, as this should also in my mind be in a state of near permanent backwardation?

Topic Author
Rob Bertram
Posts: 809
Joined: Mon May 05, 2014 12:15 pm

Re: Should I use margin to buy a balanced fund?

Post by Rob Bertram » Thu Dec 07, 2017 12:20 pm

Yeah, the 2-year note futures have been fairly stagnant. I can check my history, but I believe that they are flat or at a slight loss year to date. Note that they are doing exactly what I want: providing an uncorrelated asset to the portfolio with a positive expected return. And, I agree, the portfolio is doing well which is what should be the ultimate focus.

An inverted yield curve is definitely a concern for a leveraged portfolio. It has happened, but it is a relatively rare event. Outside of keeping leverage within reason, I don't have a portfolio mitigation strategy. I don't want to abandon my total-market framework in order to add assets that are only relevant for a specific black swan event and create drag on the portfolio the rest of the time. I understand your original goal for including VIX futures in your portfolio, but it's not an approach that I would do. As you point out, VIX futures almost always in contango which makes it a losing choice for a buy-and-hold portfolio.

For anyone interested, there is a nice animated graph of the treasury yield curve here: http://stockcharts.com/freecharts/yieldcurve.php

kassad
Posts: 1
Joined: Fri Feb 02, 2018 4:22 am

Re: Should I use margin to buy a balanced fund?

Post by kassad » Fri Feb 02, 2018 8:48 am

First of all: Kudos to you, Rob for doing this, and opening a thread about it!

Lately I have been thinking about doing the same thing (okay not exactly the same, but something similar), using futures to lever up a balanced portfolio. I haven't yet read through the entire 19 pages of this thread, but I have seen you are also using QUANDL as data source for your futures backtesting. Me too, and I have a problem with it. I am curious how you might have addressed this?

I have noticed that when I compare the buy and hold performance of the SP500 index (an ETF or US Large Cap in portfoliovisualizer) to the continious SP500 future of the CHRIS database on quandl, they follow each other nicely. However when comparing the same for US long term treasuries (the etf TLT or the cash bond index Long Term Treasuries in portfoliovisualizer) and the continious 30 year bond futures on Quandl, the continious futures vastly underperform (this is true for all the cash bond indexes vs their continious bond futures counterparts that I have tried). To the point that holding the bond futures in the portfolio aren't doing their 'job' well enough. I and up with a few percent more drawdown, and nearly half the CAGR when I put in the 40/60 stock/treasuries futures portfolio on a 1× leverage in amibroker (1982-2017) then what portfolivisualizer works out for the 40/60 index/ETF counterparts. This is a rather dissapointing simulation result.

I have to notice that the continious futures aren't rolled over based on volume or open interest, but are based on spot month calculations (held until expiry) which obviously results in a drag on the performance of futures contracts that are in contango. There is also a huge gap in all the continious bond futures on quandl I have checked from the 21st of Dec 1999 to the 22nd of Dec 1999 (110.84 -> 91.09 in the case of the 30 year treasuries !!!!!).

Did you notice the same problem? How have you gotten round it in your backtesting? Or you don't use the free continious futures at all, and have constructed your own from the individual historical futures using your own rollover rules (I have found various scripts in R and Python that do this, but I myself am not experienced in either programming languages) ?

cavemank
Posts: 1
Joined: Fri Aug 04, 2017 2:09 pm

Re: Should I use margin to buy a balanced fund?

Post by cavemank » Thu Feb 08, 2018 11:47 am

Rob,

I'm really curious as to how your portfolio is doing right now, with bonds and stocks both down a bit.

Finance Doc
Posts: 12
Joined: Thu Jan 30, 2014 8:23 pm

Re: Should I use margin to buy a balanced fund?

Post by Finance Doc » Mon Feb 12, 2018 12:17 am

Rob Bertram wrote:
Fri Dec 02, 2016 6:46 pm
That is a very good question, frejd. Here are my observations:
  • Don't let the pursuit of perfect be the enemy of good: At the core, the model is a 40% total-stock / 60% intermediate-treasury portfolio leveraged to taste. Transforming the bonds from intermediate to short-term was an exercise of keeping the same Sharpe ration and improving the Sortino ratio so as to reduce margin call. There are less complex portfolios with very similar risk/return profiles that you can achieve with a lower starting dollar amount. For example, a 10% stock / 90% short-term treasury portfolio has virtually the same risk profile as 10% stock/ 5% long-term / 85% short-term treasuries (Portfolio Visualizer link). With about $40k, one could get an e-mini S&P contract and 4x 2-year treasury contract. That would be close enough. For lower amounts, you could use leveraged ETFs for stocks (UPRO is 3x S&P 500) and 2-year treasury futures.
  • You don't need to do the insane leverage ratios that I've chosen: I would encourage people just starting out to begin with something much lower as they haven't truly experienced a bear market to understand their risk tolerance.
  • Remember the fundamentals: The top three factors for building wealth are savings rate, time, and keeping costs low. Master those first, then come back to asset allocation and risk/leverage.
I am contemplating doing something similar to what you've done and came across this discussion. A few questions for you:

1) From my cursory reading, you seem to have settled on a 10/90 futures split between s and p 500 / 2 year treasuries? Any other takeaway points to your method? I'm not a market timer at all but starting out with this approach in a market with high stock valuation and low interest rates makes me think twice.

2) Regarding your answer above, 1 e-mini and 4 treasury futures is 530,000 notional value. I don't know the exact details, is maintenance margin for that around 10 to 15 thousand? Leaving a loss of only 25 to 30 thousand to wipe out the position? Max loss maybe as low as 4.7%? I know backtesting predicts that won't happen, but they say, "Past results are no guarantee of future performance" for a reason.

Any and all advice regarding your approach would be appreciated.

long_gamma
Posts: 347
Joined: Mon Sep 14, 2015 4:13 pm

Re: Should I use margin to buy a balanced fund?

Post by long_gamma » Mon Feb 12, 2018 1:18 pm

Finance Doc wrote:
Mon Feb 12, 2018 12:17 am
2) Regarding your answer above, 1 e-mini and 4 treasury futures is 530,000 notional value. I don't know the exact details, is maintenance margin for that around 10 to 15 thousand? Leaving a loss of only 25 to 30 thousand to wipe out the position? Max loss maybe as low as 4.7%? I know backtesting predicts that won't happen, but they say, "Past results are no guarantee of future performance" for a reason.
Your notional value calculation looks wrong. 2 year treasury note face amount is $200,000. At current values of 2 year note and ES emini, notional value is close to $1 million.

I am all for leverage, but the leverage applied here is crazy for my taste.
"Everyone has a plan 'till they get punched in the mouth." --Mike Tyson

jw50
Posts: 19
Joined: Thu Jul 27, 2017 2:00 pm

Re: Should I use margin to buy a balanced fund?

Post by jw50 » Mon Feb 12, 2018 3:18 pm

Too many posts to read through.

For simpleton like me, the two main considerations are:
-the differential between the expected return and the margin interest rate
-the potential of margin call, been a forced seller

Therefore I will have only stocks in the portion of my portfolio bought with margin; and use maximum of 20% margin.

Finance Doc
Posts: 12
Joined: Thu Jan 30, 2014 8:23 pm

Re: Should I use margin to buy a balanced fund?

Post by Finance Doc » Mon Feb 12, 2018 5:48 pm

long_gamma wrote:
Mon Feb 12, 2018 1:18 pm
Finance Doc wrote:
Mon Feb 12, 2018 12:17 am
2) Regarding your answer above, 1 e-mini and 4 treasury futures is 530,000 notional value. I don't know the exact details, is maintenance margin for that around 10 to 15 thousand? Leaving a loss of only 25 to 30 thousand to wipe out the position? Max loss maybe as low as 4.7%? I know backtesting predicts that won't happen, but they say, "Past results are no guarantee of future performance" for a reason.
Your notional value calculation looks wrong. 2 year treasury note face amount is $200,000. At current values of 2 year note and ES emini, notional value is close to $1 million.

I am all for leverage, but the leverage applied here is crazy for my taste.
Okay, I must have had the specs wrong. Even worse.

bmritz
Posts: 35
Joined: Sun Nov 29, 2015 2:59 am

Re: Should I use margin to buy a balanced fund?

Post by bmritz » Mon Feb 12, 2018 6:20 pm

jw50 wrote:
Mon Feb 12, 2018 3:18 pm
Too many posts to read through.

For simpleton like me, the two main considerations are:
-the differential between the expected return and the margin interest rate
-the potential of margin call, been a forced seller

Therefore I will have only stocks in the portion of my portfolio bought with margin; and use maximum of 20% margin.
Isn't the buying power of the margin fungible though? So if you use any margin on anything on a portfolio you are effectively levering the entire portfolio? That's how I always thought of it. So the relevant measure of your first consideration would be the expected return of your entire portfolio vs the margin rate.

bmritz
Posts: 35
Joined: Sun Nov 29, 2015 2:59 am

Re: Should I use margin to buy a balanced fund?

Post by bmritz » Tue Feb 13, 2018 7:45 pm

cavemank wrote:
Thu Feb 08, 2018 11:47 am
Rob,

I'm really curious as to how your portfolio is doing right now, with bonds and stocks both down a bit.
I was also curious how this portfolio performed during the recent “correction.”

Last I could follow, the portfolio was: 10% S&P 500, 5% Long Term Treasuries, 85% Short Term Treasuries Leveraged 26x via futures.

Returns between close of Jan. 26 and close of Feb. 8:

Code: Select all

VOO (Vanguard S&P 500 ETF):                  -10.09%  (Jan 26: 263.36 Feb 8: 236.79)
VGLT (Vanguard Long-Term Treasury ETF):      -3.77%   (Jan 26: 75.88  Feb 8: 73.02)
VGSH (Vanguard Short-Term Treasury ETF):     +0.00%   (Jan 26: 60.09  Feb 8: 60.09)
By naively multiplying the portfolio weights by the returns, the total unlevered portfolio decreased by 1.20%.
Levered 26x, the account funds decreased by 31.13%.

Wild ride indeed.

For a little context, excluding dividends, YTD the unlevered portfolio is down 0.59%, while the levered portfolio is down 15.42%.

95% of that 15.42% loss came from decreased NAV in the bond portion of the portfolio. This thing is sensitive to interest rates.

jw50
Posts: 19
Joined: Thu Jul 27, 2017 2:00 pm

Re: Should I use margin to buy a balanced fund?

Post by jw50 » Thu Feb 15, 2018 5:52 am

An interesting article on leverage:
http://www.morningstar.com/articles/849 ... erage.html

long_gamma
Posts: 347
Joined: Mon Sep 14, 2015 4:13 pm

Re: Should I use margin to buy a balanced fund?

Post by long_gamma » Thu Feb 15, 2018 7:14 am

I have problem with his initial assumptions.
Real-World Costs
Of course …money is never free. I then raided the Federal Reserve Bank of St. Louis’s database (FRED), downloaded the monthly federal-funds interest rate, and simulated three borrowing costs. I permitted institutions to borrow at the fed-funds rate plus 2 percentage points; larger retail investors ($1 million), at fed-funds plus 4 points; and small retail buyers at fed-funds plus 6 points. (I couldn’t find a database that listed historical margin-account interest rates, but based on current quotes, those assumptions seem roughly correct.)
He has two links (FRED & IB) which doesn't justify the rates he is quoting.

FRED rate 1.42%
IB rates varies from 2.04% to 2.92%

So for very small retail buyer he is assuming 1.42%+6% = 7.42% which is nowhere near 2.92%.

There is other option, retail investors can borrow using futures and options at much lower rate. Implicit cost of borrowing using futures and options is less than 2.5%.

It doesn't take genius to figure out, when expected returns are around 6 to 7% and if one is borrowing at 7.42% then leverage doesn't work.
"Everyone has a plan 'till they get punched in the mouth." --Mike Tyson

n00b590
Posts: 33
Joined: Mon Dec 04, 2017 3:35 pm

Re: Should I use margin to buy a balanced fund?

Post by n00b590 » Thu Feb 15, 2018 4:17 pm

long_gamma wrote:
Thu Feb 15, 2018 7:14 am
I have problem with his initial assumptions.
Real-World Costs
Of course …money is never free. I then raided the Federal Reserve Bank of St. Louis’s database (FRED), downloaded the monthly federal-funds interest rate, and simulated three borrowing costs. I permitted institutions to borrow at the fed-funds rate plus 2 percentage points; larger retail investors ($1 million), at fed-funds plus 4 points; and small retail buyers at fed-funds plus 6 points. (I couldn’t find a database that listed historical margin-account interest rates, but based on current quotes, those assumptions seem roughly correct.)
He has two links (FRED & IB) which doesn't justify the rates he is quoting.

FRED rate 1.42%
IB rates varies from 2.04% to 2.92%

So for very small retail buyer he is assuming 1.42%+6% = 7.42% which is nowhere near 2.92%.

There is other option, retail investors can borrow using futures and options at much lower rate. Implicit cost of borrowing using futures and options is less than 2.5%.

It doesn't take genius to figure out, when expected returns are around 6 to 7% and if one is borrowing at 7.42% then leverage doesn't work.
+1

redstar
Posts: 44
Joined: Thu Jul 13, 2017 11:15 pm

Re: Should I use margin to buy a balanced fund?

Post by redstar » Mon Apr 23, 2018 11:31 pm

Rob Bertram wrote:
Tue Apr 14, 2015 12:03 pm
I was thinking about how to start a leveraged portfolio with about $10k (which is the minimum to open a margin account). Do you plan on keeping the 10/90 asset allocation as you continue to contribute cash, or is it only a transitional point due to the size of the futures contract? If you do plan on transitioning, would you be willing to share your plan?
Rob, thanks for this thread. It’s fascinating. Do you have estimates on the portfolio that would make sense to start using low amounts of leverage with a small amount of capital? I’m not sure I would do it, but I’m curious what it looks like.

Topic Author
Rob Bertram
Posts: 809
Joined: Mon May 05, 2014 12:15 pm

Re: Should I use margin to buy a balanced fund?

Post by Rob Bertram » Mon Oct 01, 2018 8:32 am

redstar wrote:
Mon Apr 23, 2018 11:31 pm
Rob, thanks for this thread. It’s fascinating. Do you have estimates on the portfolio that would make sense to start using low amounts of leverage with a small amount of capital? I’m not sure I would do it, but I’m curious what it looks like.
My apologies for a slow response. I have been busy with life and did not have free time to spend on the forums. For people starting to invest and are in the accumulation phase, the easiest thing to do is get the UPRO ETF. It is designed to represent 3x the daily value of the S&P500 and is re-leveraged daily. Due to volatility, this gives around 2x the returns of the S&P500 but you get 3x the volatility. That is what I have in my Roth IRA at Vanguard since 2014, and it has about tripled since then.

WhiteMaxima
Posts: 1927
Joined: Thu May 19, 2016 5:04 pm

Re: Should I use margin to buy a balanced fund?

Post by WhiteMaxima » Mon Oct 01, 2018 10:40 am

don't use margin loan. basically it is kind of leveraged investment at a high interest.

garlandwhizzer
Posts: 2380
Joined: Fri Aug 06, 2010 3:42 pm

Re: Should I use margin to buy a balanced fund?

Post by garlandwhizzer » Mon Oct 01, 2018 12:17 pm

WhiteMaxima wrote:

don't use margin loan. basically it is kind of leveraged investment at a high interest.
1+

Taking out a loan and leveraging up to buy a volatile asset is a high risk strategy that may work if the market is kind to you or may be a complete disaster if it is not. If the market collapses balanced funds will also suffer big time and the need to make loan payments plus margin calls can hit you at precisely the time of greatest pain. This strategy has the potential to destroy your investing future not to mention the emotional agony it can create.

Garland Whizzer

EfficientInvestor
Posts: 162
Joined: Thu Nov 01, 2018 7:02 pm

Re: Should I use margin to buy a balanced fund?

Post by EfficientInvestor » Sun Nov 04, 2018 6:09 pm

Rob Bertram wrote:
Mon Oct 01, 2018 8:32 am
For people starting to invest and are in the accumulation phase, the easiest thing to do is get the UPRO ETF. It is designed to represent 3x the daily value of the S&P500 and is re-leveraged daily. Due to volatility, this gives around 2x the returns of the S&P500 but you get 3x the volatility. That is what I have in my Roth IRA at Vanguard since 2014, and it has about tripled since then.
Rob - Thanks for starting the thread. I have not taken the time to read through all 19 pages, so my apologies if my question has already been covered. Have you considered using a blend of UPRO or TQQQ for 3X stock exposure and then combining with a 3X bond fund (TMF) and perhaps a 3X gold fund (UGLD)? My research shows that using a 35/55/10 split of TQQQ/TMF/UGLD and rebalancing at the end of each year would have resulted in a ~24% annualized return with a max drawdown of -58% since the beginning of 1987. Over that same time period, the S&P 500 had an annualized return of 10.4% with a max drawdown of -51%. So you get over 2 times the annualized return with similar downside risk.
Last edited by EfficientInvestor on Thu Feb 21, 2019 9:20 pm, edited 1 time in total.

desafinado
Posts: 154
Joined: Wed Dec 28, 2016 3:14 am

Re: Should I use margin to buy a balanced fund?

Post by desafinado » Sun Dec 30, 2018 5:35 pm

Rob, any update given rising rates and recent stock market volatility?

Benjamin Buffett
Posts: 86
Joined: Sun Nov 25, 2018 2:20 pm
Location: Indiana

Re: Should I use margin to buy a balanced fund?

Post by Benjamin Buffett » Sun Dec 30, 2018 5:41 pm

I think of buying on margin as taking a time bomb into an investment account. If the margin is called, the debt can cause a loss, in some cases a substantial loss. Mortgage investors thought it would be a good idea to buy while being heavily leveraged......it was not a good move: https://youtu.be/bx_LWm6_6tA I would not recommend buying on margin, and personally avoid potential debt loss like the plague. The Great Depression was in part caused by margin trading.

Short selling I would only do if I covered my call with a long position at the same time, but even then I would expect to take a commission loss if I cant time the market well enough to make more than my commission costs. I don't expect that I can do this reliably enough to mess with covered calls and other forms of speculation.

HEDGEFUNDIE
Posts: 3094
Joined: Sun Oct 22, 2017 2:06 pm

Re: Should I use margin to buy a balanced fund?

Post by HEDGEFUNDIE » Sun Dec 30, 2018 5:47 pm

desafinado wrote:
Sun Dec 30, 2018 5:35 pm
Rob, any update given rising rates and recent stock market volatility?
Looks like UPRO had a 50% drawdown from top to bottom compared to 20% for VOO. Sometimes the volatility decay can work in your favor!

sawvp
Posts: 6
Joined: Sat Mar 09, 2019 2:53 pm

Re: Should I use margin to buy a balanced fund?

Post by sawvp » Sat Mar 09, 2019 3:16 pm

kassad wrote:
Fri Feb 02, 2018 8:48 am
First of all: Kudos to you, Rob for doing this, and opening a thread about it!

Lately I have been thinking about doing the same thing (okay not exactly the same, but something similar), using futures to lever up a balanced portfolio. I haven't yet read through the entire 19 pages of this thread, but I have seen you are also using QUANDL as data source for your futures backtesting. Me too, and I have a problem with it. I am curious how you might have addressed this?

I have noticed that when I compare the buy and hold performance of the SP500 index (an ETF or US Large Cap in portfoliovisualizer) to the continious SP500 future of the CHRIS database on quandl, they follow each other nicely. However when comparing the same for US long term treasuries (the etf TLT or the cash bond index Long Term Treasuries in portfoliovisualizer) and the continious 30 year bond futures on Quandl, the continious futures vastly underperform (this is true for all the cash bond indexes vs their continious bond futures counterparts that I have tried). To the point that holding the bond futures in the portfolio aren't doing their 'job' well enough. I and up with a few percent more drawdown, and nearly half the CAGR when I put in the 40/60 stock/treasuries futures portfolio on a 1× leverage in amibroker (1982-2017) then what portfolivisualizer works out for the 40/60 index/ETF counterparts. This is a rather dissapointing simulation result.

I have to notice that the continious futures aren't rolled over based on volume or open interest, but are based on spot month calculations (held until expiry) which obviously results in a drag on the performance of futures contracts that are in contango. There is also a huge gap in all the continious bond futures on quandl I have checked from the 21st of Dec 1999 to the 22nd of Dec 1999 (110.84 -> 91.09 in the case of the 30 year treasuries !!!!!).

Did you notice the same problem? How have you gotten round it in your backtesting? Or you don't use the free continious futures at all, and have constructed your own from the individual historical futures using your own rollover rules (I have found various scripts in R and Python that do this, but I myself am not experienced in either programming languages) ?
I've been doing a fair bit of research as well and have found the same problem. I haven't been able to find answers as to why, but the return for treasury futures is very different than when I look at regular treasury returns in PortfolioVisualizer. For a simple look, you can go to MacroTrends and see that since 1991, the average return for the 2 year is .5% a year, with many, many down years. After you factor in the cost of slippage and contango, which has been estimated to be .4-.5%, there's nothing left. If you've found a way around this, or even an explanation, it would be great to hear about it.

Initially, the math appears to work wonderfully well for this 10/90 stock/treasury split Rob has set up, but leveraging a treasury etf with a margin loan is too expensive right now (IB rates have gone up). Leveraging with futures is nearly free, but the lack of real return paints a very different picture for the portfolio. It would be great to hear from Rob about how this is panning out in reality. My research tells me that 2 year treasury futures have lost money every single calendar year since he started this thing--while the SHY ETF has made money every year. Clearly, I still have a lot to learn about treasuries...

Beliavsky
Posts: 932
Joined: Sun Jun 29, 2014 10:21 am

Re: Should I use margin to buy a balanced fund?

Post by Beliavsky » Sun Mar 10, 2019 3:40 pm

sawvp wrote:
Sat Mar 09, 2019 3:16 pm
I've been doing a fair bit of research as well and have found the same problem. I haven't been able to find answers as to why, but the return for treasury futures is very different than when I look at regular treasury returns in PortfolioVisualizer. For a simple look, you can go to MacroTrends and see that since 1991, the average return for the 2 year is .5% a year, with many, many down years. After you factor in the cost of slippage and contango, which has been estimated to be .4-.5%, there's nothing left. If you've found a way around this, or even an explanation, it would be great to hear about it.

Initially, the math appears to work wonderfully well for this 10/90 stock/treasury split Rob has set up, but leveraging a treasury etf with a margin loan is too expensive right now (IB rates have gone up). Leveraging with futures is nearly free, but the lack of real return paints a very different picture for the portfolio. It would be great to hear from Rob about how this is panning out in reality. My research tells me that 2 year treasury futures have lost money every single calendar year since he started this thing--while the SHY ETF has made money every year. Clearly, I still have a lot to learn about treasuries...
You are probably making the same mistakes discussed in the thread Understanding using treasury futures for leverage to implement risk parity.

Topic Author
Rob Bertram
Posts: 809
Joined: Mon May 05, 2014 12:15 pm

Re: Should I use margin to buy a balanced fund?

Post by Rob Bertram » Thu Aug 08, 2019 5:24 pm

sawvp wrote:
Sat Mar 09, 2019 3:16 pm
've been doing a fair bit of research as well and have found the same problem. I haven't been able to find answers as to why, but the return for treasury futures is very different than when I look at regular treasury returns in PortfolioVisualizer. For a simple look, you can go to MacroTrends and see that since 1991, the average return for the 2 year is .5% a year, with many, many down years. After you factor in the cost of slippage and contango, which has been estimated to be .4-.5%, there's nothing left. If you've found a way around this, or even an explanation, it would be great to hear about it.

Initially, the math appears to work wonderfully well for this 10/90 stock/treasury split Rob has set up, but leveraging a treasury etf with a margin loan is too expensive right now (IB rates have gone up). Leveraging with futures is nearly free, but the lack of real return paints a very different picture for the portfolio. It would be great to hear from Rob about how this is panning out in reality. My research tells me that 2 year treasury futures have lost money every single calendar year since he started this thing--while the SHY ETF has made money every year. Clearly, I still have a lot to learn about treasuries...
You are correct. Leveraging a short-term treasury on average nets zero real return or possibly some loss due to market movements. Long-term treasuries do have a positive return after financing costs, but they are a relatively small percentage of the portfolio.

The purpose of the short-term treasuries is not for return but to be an uncorrelated asset to compliment my equity allocation. It allows me to effectively leverage stocks by 2.6x and not go bust when speculators panic and the market tanks.

I have recently been assessing the need for short-term treasuries in my portfolio mainly for purposes of reducing complexity. I do not have an automatic way to auto-balance, so I mainly log in to roll contracts. Since enjoying life is way more fun than manually entering trades, I am thinking about reducing the portfolio to 2x stocks and no bonds. The risk profile is about the same (slightly worse), the return is slightly worse, but my monthly contributions would prevent a margin call. This change means I only have to log in 4 times a year to roll contracts instead of 8 times.

Topic Author
Rob Bertram
Posts: 809
Joined: Mon May 05, 2014 12:15 pm

Re: Should I use margin to buy a balanced fund?

Post by Rob Bertram » Thu Aug 08, 2019 5:30 pm

If anyone has any pressing questions, please send me a private message as I do not check the forum regularly. Happy investing!

EfficientInvestor
Posts: 162
Joined: Thu Nov 01, 2018 7:02 pm

Re: Should I use margin to buy a balanced fund?

Post by EfficientInvestor » Mon Aug 12, 2019 6:49 am

sawvp wrote:
Sat Mar 09, 2019 3:16 pm
kassad wrote:
Fri Feb 02, 2018 8:48 am
First of all: Kudos to you, Rob for doing this, and opening a thread about it!

Lately I have been thinking about doing the same thing (okay not exactly the same, but something similar), using futures to lever up a balanced portfolio. I haven't yet read through the entire 19 pages of this thread, but I have seen you are also using QUANDL as data source for your futures backtesting. Me too, and I have a problem with it. I am curious how you might have addressed this?

I have noticed that when I compare the buy and hold performance of the SP500 index (an ETF or US Large Cap in portfoliovisualizer) to the continious SP500 future of the CHRIS database on quandl, they follow each other nicely. However when comparing the same for US long term treasuries (the etf TLT or the cash bond index Long Term Treasuries in portfoliovisualizer) and the continious 30 year bond futures on Quandl, the continious futures vastly underperform (this is true for all the cash bond indexes vs their continious bond futures counterparts that I have tried). To the point that holding the bond futures in the portfolio aren't doing their 'job' well enough. I and up with a few percent more drawdown, and nearly half the CAGR when I put in the 40/60 stock/treasuries futures portfolio on a 1× leverage in amibroker (1982-2017) then what portfolivisualizer works out for the 40/60 index/ETF counterparts. This is a rather dissapointing simulation result.

I have to notice that the continious futures aren't rolled over based on volume or open interest, but are based on spot month calculations (held until expiry) which obviously results in a drag on the performance of futures contracts that are in contango. There is also a huge gap in all the continious bond futures on quandl I have checked from the 21st of Dec 1999 to the 22nd of Dec 1999 (110.84 -> 91.09 in the case of the 30 year treasuries !!!!!).

Did you notice the same problem? How have you gotten round it in your backtesting? Or you don't use the free continious futures at all, and have constructed your own from the individual historical futures using your own rollover rules (I have found various scripts in R and Python that do this, but I myself am not experienced in either programming languages) ?
I've been doing a fair bit of research as well and have found the same problem. I haven't been able to find answers as to why, but the return for treasury futures is very different than when I look at regular treasury returns in PortfolioVisualizer. For a simple look, you can go to MacroTrends and see that since 1991, the average return for the 2 year is .5% a year, with many, many down years. After you factor in the cost of slippage and contango, which has been estimated to be .4-.5%, there's nothing left. If you've found a way around this, or even an explanation, it would be great to hear about it.

Initially, the math appears to work wonderfully well for this 10/90 stock/treasury split Rob has set up, but leveraging a treasury etf with a margin loan is too expensive right now (IB rates have gone up). Leveraging with futures is nearly free, but the lack of real return paints a very different picture for the portfolio. It would be great to hear from Rob about how this is panning out in reality. My research tells me that 2 year treasury futures have lost money every single calendar year since he started this thing--while the SHY ETF has made money every year. Clearly, I still have a lot to learn about treasuries...
I'm thinking that this MacroTrends data is not a total return index. To accurately reflect what you are doing in PortfolioVisualizer, I believe you want to look at a Total Return Index. Take a look at the difference between the 2-year indexes that ICE offers at the links below. The non total return index is similar to the MacroTrends chart you referenced. The total return index chart looks more like the returns of SHY.

ICE 2-year treasury futures index - https://www.nyse.com/quote/index/USTTWO
ICE 2-year treasury futures total return index - https://www.nyse.com/quote/index/USTTWOT

When using futures to replicate SHY, you would put a small amount of capital in the futures contract and then invest the rest in T-bills that receive the risk-free rate. This is essentially what the Total Return index assumes.

Let me know if you agree with this analysis.

sawvp
Posts: 6
Joined: Sat Mar 09, 2019 2:53 pm

Re: Should I use margin to buy a balanced fund?

Post by sawvp » Tue Aug 13, 2019 7:24 pm

Thanks for the links. I've always understood total return to include dividends, which you don't receive on a futures contract. I could very well be wrong though--let me know if I'm mistaken. As you point out, the interest earned in cash is definitely a bonus, though I'm thinking that number could be dropping hard and fast in the coming weeks/months.

Since originally posting the question I bought SHY as well as 1 futures contract (ZT) to see how similar they perform. I've come to realize the futures contract is a different product and can't be expected to mirror SHY. It's betting on a future price. Some days I make money in SHY and lose it in ZT and vice versa. Or some days, SHY is flat and ZT does great. I have to agree with Rob's last statement, that it's more of an insurance policy against a major draw down in risk assets--though it's been profitable as of late and I wished I bought more than one contract. I'm underweight in stocks right now, but even once I ramp back up, I'll probably always hold some treasury contracts as cheap insurance. You do have to roll them every three months, but this takes about 5 minutes. If there's anything I've gained from this thread--it's been the proper introduction to ZT. I am a big fan. THANK YOU ROB.

rhe
Posts: 51
Joined: Sun Feb 26, 2017 2:10 am

Re: Should I use margin to buy a balanced fund?

Post by rhe » Wed Aug 14, 2019 2:31 am

Rob Bertram wrote:
Thu Aug 08, 2019 5:24 pm
I have recently been assessing the need for short-term treasuries in my portfolio mainly for purposes of reducing complexity. I do not have an automatic way to auto-balance, so I mainly log in to roll contracts. Since enjoying life is way more fun than manually entering trades, I am thinking about reducing the portfolio to 2x stocks and no bonds. The risk profile is about the same (slightly worse), the return is slightly worse, but my monthly contributions would prevent a margin call. This change means I only have to log in 4 times a year to roll contracts instead of 8 times.
I've basically changed over to using eurodollar futures (and equivalents in other currencies) for this reason. These are cash settled, and if you have a two year strip of eurodollars and forget to do anything for a quarter, all that happens is that you end up with a 1.75 year strip of eurodollars. Now I can just log in whenever the thought occurs to me, rather than worrying about rolling on certain specific days. Roll costs overall are also lower, once you take into account that you're only trading 1/8 of your contracts each quarter.

RandomWord
Posts: 32
Joined: Tue Jun 18, 2019 1:12 pm

Re: Should I use margin to buy a balanced fund?

Post by RandomWord » Wed Aug 14, 2019 12:20 pm

sawvp wrote:
Tue Aug 13, 2019 7:24 pm
Since originally posting the question I bought SHY as well as 1 futures contract (ZT) to see how similar they perform. I've come to realize the futures contract is a different product and can't be expected to mirror SHY. It's betting on a future price. Some days I make money in SHY and lose it in ZT and vice versa. Or some days, SHY is flat and ZT does great. I have to agree with Rob's last statement, that it's more of an insurance policy against a major draw down in risk assets--though it's been profitable as of late and I wished I bought more than one contract. I'm underweight in stocks right now, but even once I ramp back up, I'll probably always hold some treasury contracts as cheap insurance. You do have to roll them every three months, but this takes about 5 minutes. If there's anything I've gained from this thread--it's been the proper introduction to ZT. I am a big fan. THANK YOU ROB.
You're accounting for the implied cost of leverage in ZT, right? 100k of notional value in ZT is going to do much worse than 100k in SHY, because the ZT is effectively borrowing money and paying interest. Also SHY is a wider range of duration in its treasuries.

My understanding is that the "normal" state of affairs, with a normal yield curve, is that ZT earns the difference between the very short term rates and the 2-year rate. You borrow short and lend long, like a bank. The interest rate movements are just noise on top of that. It's weird now because the yield curve is inverted, so in theory it's paying more to borrow than what it earns from lending, but "everyone" expects the short term rates to drop, so the price movements are dominated by news that affects when/how fast the short term rates will drop.

Post Reply