Should I use margin to buy a balanced fund?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Topic Author
Rob Bertram
Posts: 859
Joined: Mon May 05, 2014 12:15 pm

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

Post by Rob Bertram »

Spec7re wrote:Rob, when did you decide to jump to 26x leverage? You started at 6x. Do you think 26x is safe?

Btw, I'm holding 3 2yr futures with about 80k nominal now, about 9x leverage.
I made the decision in early-mid August after reading a thread created by Nisiprius where he showed that longer-term bonds behave like a leveraged version of short-term bonds in a balanced portfolio. I posted about it on the 14th and executed it on August 26th when it was time to roll futures.

Nisiprius has a strong point that long-term bonds behave like leveraged short-term bonds when looking at a portfolio behavior on an annual basis. On a daily basis, I feel that long-term bonds add an uncorrelated volatility to the portfolio that offsets stocks which is why I still have some long-term bonds in my portfolio.

The nice thing about short-term bonds is that they recover much faster from rate increases than long-term (duration). I mainly need to watch the 2-year yield and take action if it drops below my financing rate.

And to be clear, I changed the asset allocation when I changed leverage. I went from 6x 40 % stocks/30% long-term treasuries/30% intermediate treasuries to 26x 10% stock / 5% long-term treasures / 85% short-term treasuries.

So to answer your question, I believe that 26x is acceptable for my situation and risk tolerance. (Nothing is ever "safe.") I am adding $3k to the portfolio monthly. I plan my portfolio to have a historical worst case of 70-80% loss. The future might be worse than anything in history, so I might lose 100% or 120%. As the portfolio grows relative to my monthly deposits, I will reduce my leverage.

One last thing that I will say is that with a higher leverage ratio, I expect more alpha from volatility harvesting. Even though both 6x and 26x portfolios have similar CAPM risk/return expectations, I believe that the 26x portfolio will see slightly better returns.
Topic Author
Rob Bertram
Posts: 859
Joined: Mon May 05, 2014 12:15 pm

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

Post by Rob Bertram »

Daryl wrote:Clearly this is why I shouldn't be trading Treasury futures!

If I buy bonds at the spot price, and yields increase, the price of that bond decreases.
Remember that the same happens if you own a bond fund. The NAV of the fund decreases when yields increase.
Daryl wrote:If I buy Treasury futures, and yields increase (I understand that I wouldn't not benefit from this yield), wouldn't the value of that contract decrease? Does someone who is long treasury futures want interest rates to rise or fall?
You are mistaken here. You do benefit from the increased yield. For starters, as you approach the contract date of a future, the contract price approaches the spot price because there is less pro-rata dividend yield and financing. Each day, the price of the contract increases to reflect one day of yield and one less day of financing. So yes, when yields rise, the contract price drops that day. However, you do benefit from the yield as long as you continue to hold that contract.

If we ignore taxes, the main difference between holding a bond future and the actual asset is the financing cost.
Clive
Posts: 1950
Joined: Sat Jun 13, 2009 5:49 am

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

Post by Clive »

Rob Bertram wrote:I hope that you consider treasury futures as well. The cost to roll futures comes to about 5-8 basis points a year. There could be some savings over a bond expense ratio if you were getting a treasury ETF. Since this is in a Roth IRA account, you don't need to worry about taxes. However, if you ever did this in a taxable account, be aware that futures are taxed at 60% long-term/40% short-term capital gains
Ouch! (60%/40% tax rate).

In the UK we have spreadbets and at least one firm allows you to set variable margin on rolling positions i.e. tight spreads and once bought into at spot price you pay a cost of carry nightly (typically a pro-rata LIBOR+2.5% rate), but if fully funded then there's no cost of carry. Being a bet they're exempt from capital gains and income tax. Our more conventional non taxed accounts (ISA's) are one-way-locked, you can withdraw at any time without penalty but can't then return funds back into the account. You're limited to adding a maximum of around $22,000/year to such accounts.
azanon
Posts: 3142
Joined: Mon Nov 07, 2011 9:34 am

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

Post by azanon »

I would say I never borrow to invest, but then I'm almost immediately reminded.... I have a mortgage, and investments at the same time. :oops:
Topic Author
Rob Bertram
Posts: 859
Joined: Mon May 05, 2014 12:15 pm

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

Post by Rob Bertram »

To put things into perspective, a majority of returns from bonds are taxed as ordinary income which is taxed at the marginal rate. Long-term capital gains are taxed at a reduced rate whereas short-term cap gains are taxed at the marginal rate. Additionally, capital losses offset capital gains but there is a limit to the amount of capital losses that can offset ordinary income each year. So in this way, bond futures have an advantage over bonds in a taxable account because you go from 100% taxed at marginal rate to 60% at favorable +40% at marginal.

Now, stock futures are a different story. A majority of returns from stocks are long-term capital gains or qualified dividends which are taxed at the reduced rate. So using stock futures instead of a stock fund/ETF has a noticeable tax impact.
Daryl
Posts: 616
Joined: Thu May 22, 2008 9:34 am
Location: Texas

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

Post by Daryl »

Rob Bertram wrote: If I may ask, what asset allocation are you considering for your leveraged portfolio?
Before considering Futures, I was at 40/60 (Stocks/Fixed Income), using broad based index mutual funds (Total Stock / Total Bond). Total equity exposure is about $100K, so my thought was to sell the Total Stock Market Fund and use an ES contract to maintain that equity exposure.

By selling the Total Stock Market Index Fund, and buying an ES contract, I’d free up some capital to increase my fixed income holdings. Right now I'm just looking for cheap beta. I'm willing to do some work on my own, but don't anticipate making very many trades (other than quarterly rolls). I’m still in the planning stage, but I’m thinking about something like this:

Before:
100K Total Stock Market Fund; 150K Total Bond Market Fund (40/60 allocation)
After
100K ES (Notional Value); 15K Cash; 225K Total Bond Market Fund (Approximately 30/70 allocation)

I’ll read more about Treasury Futures over the weekend. I appreciate this recommendation!

I’m fairly certain that ThinkOrSwim doesn’t pay much interest on uninvested cash, which is why I was looking for ways to optimize this value. With 15K Cash and 1 ES contract, I could weather some fairly volatile times without liquidating any part of my account or making any additional trades. As long as ThinkOrSwim has somewhat similar margin policies, I should be able to maintain my futures position through any market conditions. My contingency plan, should we have a more severe correction, would be to sell some of the bond fund and rebalance into a second ES contract or shares of the Total Stock Market ETF.
Topic Author
Rob Bertram
Posts: 859
Joined: Mon May 05, 2014 12:15 pm

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

Post by Rob Bertram »

Daryl wrote:Before:
100K Total Stock Market Fund; 150K Total Bond Market Fund (40/60 allocation)
After
100K ES (Notional Value); 15K Cash; 225K Total Bond Market Fund (Approximately 30/70 allocation)

I’ll read more about Treasury Futures over the weekend. I appreciate this recommendation!
That looks like 250k notional/0k loan vs 325k/-75k loan which is 325/250 = 1.3x leverage.

I did a lot of analysis on leveraging a 40/60 portfolio. At 2.5x, the max drawdown is similar to a 60/40 unleveraged portfolio but returns were closer to 100% stocks. Your 30/70 portfolio leveraged 1.3x will likely have returns similar to 40/60 but the max drawdown will be slightly less.

How often will you rebalance? TBM pays out dividends every month. Will you have dividends reinvested or going to cash?
Daryl
Posts: 616
Joined: Thu May 22, 2008 9:34 am
Location: Texas

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

Post by Daryl »

I'd reinvest the dividends from TBM. We are only looking at a couple hundred dollars in monthly dividends, and it is easy enough to sell ETFs to increase liquidity, if the need arises.
I did a lot of analysis on leveraging a 40/60 portfolio. At 2.5x, the max drawdown is similar to a 60/40 unleveraged portfolio but returns were closer to 100% stocks. Your 30/70 portfolio leveraged 1.3x will likely have returns similar to 40/60 but the max drawdown will be slightly less.
This makes sense, given that I'm going from a fully funded equity position ($100K Total Stock Market) to 1 S&P emini (Notional value, about $100K)
Topic Author
Rob Bertram
Posts: 859
Joined: Mon May 05, 2014 12:15 pm

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

Post by Rob Bertram »

I thought that I would talk a little more about why I switched to include short-term bonds. Bond duration is an amazing thing. A very nice aspect of the short-term(duration) treasuries is that they have a relatively quick recovery time when rates rise. There is a shorter period of time where your return is negative compared to a longer-term bond. You might not hit a break-even point for several months, but you get a positive return slightly faster than a longer-term bond. This is important when using leverage as we are magnifying the return around zero.

For example, here is the ETF SHY as a proxy for 2-year bonds:
Image
Image

Since I only buy/re-balance/re-leverage when my leverage ratio is below target, holding a significant portion of short-term bonds allows for more volatility harvesting. When rates rose at the end of April/beginning of May, my 40/30/30 portfolio lost so much that I couldn't re-leverage until now. If I had shorter-term bonds, then there would have been some more buying opportunities, possibly some in August when stocks had their correction.
Topic Author
Rob Bertram
Posts: 859
Joined: Mon May 05, 2014 12:15 pm

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

Post by Rob Bertram »

ogd wrote:P.S.: I know it's your thread, but you might consider starting a new one. The pages and pages of good discussion on balanced allocation efficiency have sort of taken an identity of their own :)
I agree. I will talk with the admins about splitting the thread and moving the Martingale treasury discussion to it's own topic.
User avatar
LadyGeek
Site Admin
Posts: 95696
Joined: Sat Dec 20, 2008 4:34 pm
Location: Philadelphia
Contact:

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

Post by LadyGeek »

^^^ The discussion has been split into a new topic here: Should I use margin to buy a balanced fund? [Martingale treasury]
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
long_gamma
Posts: 420
Joined: Mon Sep 14, 2015 4:13 pm

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

Post by long_gamma »

Interesting thread.

Haven't gone thru' this thread completely. Have you tracked your positions so far?

Did you stressed your portfolio to certain scenario's like flash crash or 3 sigma events?
"Everyone has a plan 'till they get punched in the mouth." --Mike Tyson
lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 10:55 pm

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

Post by lack_ey »

long_gamma wrote:Interesting thread.

Haven't gone thru' this thread completely. Have you tracked your positions so far?

Did you stressed your portfolio to certain scenario's like flash crash or 3 sigma events?
I think this is it, from the last page (a few months ago).
Rob Bertram wrote:I'd say 50 bps is a fair approximation for the borrowing premium. The implied financing rate on the e-mini S&P 500 looks to be around 44 bps which might be about 30-35 bps above the FFR (13-14 bps) or about 42 bps above the 3-month T-bill (2 bps).

For those interested, I updated my spreadsheets for the past couple weeks. The links are below:
Original (using ETFs): https://docs.google.com/spreadsheets/d/ ... sp=sharing
Latest (using futures + ETFs): https://docs.google.com/spreadsheets/d/ ... sp=sharing
The OP's strategy seems to be evolving, though.
Topic Author
Rob Bertram
Posts: 859
Joined: Mon May 05, 2014 12:15 pm

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

Post by Rob Bertram »

The strategy is evolving the more I learn. The core concept comes from modern portfolio theory -- that a balanced portfolio of stocks and bonds has higher risk-adjusted returns than a single asset portfolio. I then apply leverage to dial the risk to my personal taste (CAPM).

I switched from using the total bond market ETF (and margin) to futures in order to increase my leverage beyond 2x and to reduce my borrowing costs. There is weak evidence that long-term treasuries recently have low (possibly negative) correlation to stocks, so I adjusted the bond portion of my portfolio accordingly. And, finally, there was a recent argument that duration in bonds is functionally a form of leverage (i.e. longer duration bonds behave like leveraged shorter-term bonds), so I eliminated intermediate-term bonds and adjusted my AA to only have a small portion of long-term treasuries and big portion of short-term bonds. I still believe that long-term treasuries add uncorrelated volatility to complement stocks which is why there are still some long-term treasuries in the portfolio.

Currently, the target asset allocation is 10% stocks, 5% long-term (20+ year) treasuries, and 85% short-term (2-year) treasuries. Target leverage is 26x.
long_gamma
Posts: 420
Joined: Mon Sep 14, 2015 4:13 pm

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

Post by long_gamma »

Thx. lack_ey.

Did a quick check of flash crash calculation. Stock futures dropped around 8.6%, 2 year notes rallied from 109 to 109' 13 5/32 on that day and long bond rallied about 4 points. He has about 3 ES, 1 ZB and 15 ZT according to his worksheet.

If similar flash crash happens today,
  • ES will loose $26570
    ZB will gain $4000 and
    ZT will gain $12675
I am not sure what is the collateral in the account, but kind of pleasantly surprised by the result
"Everyone has a plan 'till they get punched in the mouth." --Mike Tyson
Topic Author
Rob Bertram
Posts: 859
Joined: Mon May 05, 2014 12:15 pm

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

Post by Rob Bertram »

That was a good day to be diversified, right? I only had 2 ES contracts (and about $80k of VTI) at the time of the flash crash and sadly no ZT until later in August, even then I only picked up 10 contracts of ZT. I did have a contract each of 5- and 10-year treasury futures which also softened the blow.

The part I like about having a high leverage ratio is that small gains allow me to buy full contracts. The gains between August and now allowed me to pick up another ES contract, $17k more VTI and 5 ZT contracts. I'm a little over my target leverage, but I wanted to keep my stock/bond ratio close to the AA.

On the flip side, bonds took a huge hit in May/June, so I've been carrying a loss since then. I checked my private spreadsheet today, and I was happy to see that I was actually net positive by $700 or so. It is practically zero, but it's ahead of the "benchmark" total stock portfolio which I believe is still negative. And, of course, the market can drop again tomorrow.
long_gamma wrote:I am not sure what is the collateral in the account, but kind of pleasantly surprised by the result
Column B of the online spreadsheet shows my cash contributions which is around $125,585.79. According to Interactive Brokers, about $14.6k is sitting as cash (earning nothing), about $98k in ETFs, and another $14k sitting in a savings account earning 1% interest. The money sitting in the savings account is part of the $125k, but I'm not tracking the 1% interest in the online spreadsheet.
Day9
Posts: 1000
Joined: Mon Jun 11, 2012 6:22 pm

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

Post by Day9 »

I have a somewhat basic question regarding duration and leverage.

The yield curve is positively sloped but margin costs are fixed. I've seen in this thread that implied financing of treasury futures is about 0.25% at the moment

Here is the yield curve and the yield after subtracting financing costs:
2-year: 0.66% yield - 0.25% financing = 0.41% net
5-year: 1.45% yield - 0.25% financing = 1.20% net
10-year: 2.09% yield - 0.25% financing = 1.84% net

The longer duration bonds have more duration risk and higher yield but the idea is to use enough leverage with a shorter duration treasury to approximately match the risks of a less leveraged, higher duration treasury position. The benefits are that lower duration has more favorable correlation to stocks and recover from price drops more quickly. But it looks to me that the financing costs reduces your expected return too much from a 2-year treasury to make up for the benefits.

I'm sorry if I am not able to articulate my concern well but does this make sense? I may have missed your analysis in this thread where this is discussed so I apologize if I have done that.

[Note: Comment moved here from the spinoff thread by user]
I'm just a fan of the person I got my user name from
Topic Author
Rob Bertram
Posts: 859
Joined: Mon May 05, 2014 12:15 pm

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

Post by Rob Bertram »

You still have a positive yield after you subtract the financing cost. You just need to buy more short-term treasuries. My asset allocation has 17x more short term treasuries than long-term. At my current leverage ratio, I have about $3.3 million notional in short-term treasuries and about $160k in long-term treasuries. (And when I say long-term treasuries, I mean 20+ years.)

When you leverage a portfolio, you magnify the returns "away" from 0. By that I mean if you have a slightly positive return, leverage makes that a large positive return. On the flip side, if you have a slightly negative return, leverage turns that into a very large negative return. Because we can potentially lose everything with leverage, we need to pay special attention to how and why our portfolio can get negative returns. One might alter the quote from Fight Club to something like this: The first rule of using leverage is "don't lose money." The second rule using leverage is "don't lose money."

So Nisiprius showed that in a balanced portfolio of stocks and bonds, he could get identical long-term risk-adjusted returns with any duration of bonds and stocks if he was allowed to adjust the asset allocation and use leverage. Conceptually, we could replace all the bonds in the portfolio with short-term treasuries and just increase leverage to get the target return.

Nisiprius did long-term efficient-frontier analysis. In a shorter time frame, short-term bonds react differently to interest rate changes than long-term bonds. Short-term bonds have less interest rate sensitivity and lose less value when rates go up -- which is exactly what we want with the first rule of using leverage. As long as financing costs are below the 2-year yield, they are good to have in your leveraged portfolio.
long_gamma
Posts: 420
Joined: Mon Sep 14, 2015 4:13 pm

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

Post by long_gamma »

Rob Bertram wrote:That was a good day to be diversified, right? I only had 2 ES contracts (and about $80k of VTI) at the time of the flash crash and sadly no ZT until later in August, even then I only picked up 10 contracts of ZT. I did have a contract each of 5- and 10-year treasury futures which also softened the blow.
Flash crash I was referring to was in 2010.

https://en.wikipedia.org/wiki/2010_Flash_Crash

I think more stress for your strategy is 1994 scenario, when 2 year rallied and equities were down.

Image
"Everyone has a plan 'till they get punched in the mouth." --Mike Tyson
Topic Author
Rob Bertram
Posts: 859
Joined: Mon May 05, 2014 12:15 pm

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

Post by Rob Bertram »

long_gamma wrote:Flash crash I was referring to was in 2010.
Sorry, I'm familiar with that one. I was just thinking about the mini one we had in August.
long_gamma wrote:I think more stress for your strategy is 1994 scenario, when 2 year rallied and equities were down.
That was a brutal year, and it's one of the years that I modeled. I have a personal risk tolerance of 80% loss which was hit.
lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 10:55 pm

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

Post by lack_ey »

You, um, probably mentioned it somewhere in here, but what was the rationale behind using the 2-year as opposed to the 5-year, 10-year, or something else? Do you have any sense of the long-term differences of these approaches and potential issues in the short term as well?

With the longer-dated bonds you would use less of them, so keeping that in mind, what's the key difference here? Is there a backtested argument, a theoretical one, or what?
lee1026
Posts: 396
Joined: Sat Jun 21, 2014 7:47 pm

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

Post by lee1026 »

The argument is that shorter dated treasuries have a better sharpe ratio than longer ones. Since we are using hilariously large amounts of leverage anyway, all that matters is the sharpe ratio, not absolute return.
Topic Author
Rob Bertram
Posts: 859
Joined: Mon May 05, 2014 12:15 pm

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

Post by Rob Bertram »

That is close. My decision to change to the 2-year and some 20+ year treasuries were based on a thread started by Nisiprius. His argument as I understand it is that I can get the same long-term Sharpe ratio (risk-adjusted return) by using short-term bonds and stocks as opposed to long-term bonds and stocks. I just need to tweak the asset allocation and leverage to taste. Here's what I said a while back:
Rob Bertram wrote:Nisiprius has a thread called Does higher volatility make long-term bonds + stocks better than intermediate-term bonds + stocks?. He shows that bonds with different duration have the same risk-adjusted returns, so one can convert a stock/long-term bond portfolio to a stock/intermediate bond by adjusting the asset allocation (and possibly adding leverage).

With that in mind, I could see simplifying my portfolio from 5x 40% stock/30% long-term bond/30% intermediate-term bond to (for example) 10x 10% stock/90% short-term bond. I'm not sure if that's the correct AA, but I'll definitely spend some time pulling up historical data for 2-year treasuries.
Having posted all that, I still believe that there is a non-trivial amount of un/negative correlation between long-term treasuries and stocks in the short-term that reduce portfolio volatility which cannot be captured with short-term treasuries. That is why there are still some long-term treasuries in the portfolio.

While the long-term Sharpe ratio might be the same, different aspects of the risk profile improve when I switch to a portfolio that is heavily weighted in short-term treasuries. For example, max drawdown is more favorable for the short-term treasury portfolio. The Sortino ratio is higher. Here's an example in Portfolio Visualizer: 40/60 vs 10/5/85 (asset class, fund)

As for the leverage ratio, say that I'm comfortable with a maximum 75-80% loss. The 40/60 portfolio saw a 15% drawdown, so I was going about 5-6x with that portfolio. The 10/5/85% portfolio saw a 3% max drawdown, so I could go about 25-26x. And I am still in the "accumulation" phase, I can stay at the upper end of those leverage numbers and not sweat it.
lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 10:55 pm

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

Post by lack_ey »

Oh, okay, thanks.

That carries through when taking realistic borrowing rates into account (i.e. higher than the risk-free rate, what you get from futures)?

Also, how often do you get the diversification properties especially from long-term bonds if the plan is to unwind a trade if the carry is negative? A lot of times stocks tank after the yield curve inverts.
Topic Author
Rob Bertram
Posts: 859
Joined: Mon May 05, 2014 12:15 pm

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

Post by Rob Bertram »

Well, I'll preface this by saying that the markets are unpredictable and that there is insufficient data to make any statistically significant conclusions.

With that out of the way, I was looking at 2-year treasury futures data to see how the portfolio would hold up when borrowing rates exceed the 2-year yield. For example, 1994, 2006, and 2013. A buy-and-hold 2-year treasury futures position would have lost money but not much. Technically, 2013 ended slightly positive. I also noticed that 2-year treasury prices don't move much in a day and that there seems to be enough volatility that a little market timing would offset the losses. That inspired the Martingale market timing thread.

Correlations always change, but it's been strongly negative between stocks and long treasuries (about -0.57) over the past 5 years according to PV. That could all change tomorrow. I doubt it, but Mr. Market is full of surprises. So long-term treasuries will always be a part of the portfolio.

I don't know what the Fed will do, but I'm guessing that FFR will not rise too quickly -- maybe 2% over the next 4 years. So right now, the 2-year treasury yield is around 0.90% and 20+ treasuries are around 3.0%. It will be a while before I need to decide how to handle the carry problem. I like the idea of the Martingale strategy. That essentially turns 2-year treasuries into a permanent part of the portfolio.
Day9
Posts: 1000
Joined: Mon Jun 11, 2012 6:22 pm

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

Post by Day9 »

Rob Bertram wrote:You still have a positive yield after you subtract the financing cost. You just need to buy more short-term treasuries. My asset allocation has 17x more short term treasuries than long-term. At my current leverage ratio, I have about $3.3 million notional in short-term treasuries and about $160k in long-term treasuries. (And when I say long-term treasuries, I mean 20+ years.)

When you leverage a portfolio, you magnify the returns "away" from 0. By that I mean if you have a slightly positive return, leverage makes that a large positive return. On the flip side, if you have a slightly negative return, leverage turns that into a very large negative return. Because we can potentially lose everything with leverage, we need to pay special attention to how and why our portfolio can get negative returns. One might alter the quote from Fight Club to something like this: The first rule of using leverage is "don't lose money." The second rule using leverage is "don't lose money."

So Nisiprius showed that in a balanced portfolio of stocks and bonds, he could get identical long-term risk-adjusted returns with any duration of bonds and stocks if he was allowed to adjust the asset allocation and use leverage. Conceptually, we could replace all the bonds in the portfolio with short-term treasuries and just increase leverage to get the target return.

Nisiprius did long-term efficient-frontier analysis. In a shorter time frame, short-term bonds react differently to interest rate changes than long-term bonds. Short-term bonds have less interest rate sensitivity and lose less value when rates go up -- which is exactly what we want with the first rule of using leverage. As long as financing costs are below the 2-year yield, they are good to have in your leveraged portfolio.

Thanks I think I understand all that. My question is basically: Was the analysis that led you to switch to short term treasuries done with or without considering financing costs?

Is it true that financing costs are roughly fixed across different duration futures and thus eat up a larger percentage of the yield of shorter duration? I am wondering if it is less efficient to use greater leverage on a shorter duration instrument rather than less leverage on a longer duration one because of financing costs.
I'm just a fan of the person I got my user name from
Topic Author
Rob Bertram
Posts: 859
Joined: Mon May 05, 2014 12:15 pm

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

Post by Rob Bertram »

Day9 wrote:Thanks I think I understand all that. My question is basically: Was the analysis that led you to switch to short term treasuries done with or without considering financing costs?

Is it true that financing costs are roughly fixed across different duration futures and thus eat up a larger percentage of the yield of shorter duration? I am wondering if it is less efficient to use greater leverage on a shorter duration instrument rather than less leverage on a longer duration one because of financing costs.
I do look at financing costs. Though, my core model only used index values (no financing costs). My leverage target is based on a best guess of maximum drawdown which does include financing costs. When borrowing costs increase, leverage drifts down.

Yep. Financing costs on treasury futures are the LIBOR 3-month rate, so yes, you pay more on shorter-term treasuries per contract. Financing on S&P 500 futures (e-mini) is slightly higher. As long as financing costs stay below the 2-year yield, leveraging short-term treasuries is the clear winner. And from my analysis of periods where the yield curve inverts, leveraged short-term treasuries are less bad than long-term treasuries.
aaaaaaabbbbbbbbbb
Posts: 133
Joined: Sat Apr 11, 2015 12:34 am

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

Post by aaaaaaabbbbbbbbbb »

Here are a couple questions I have for you Rob (I apologize if they've already been answered, but I don't have time to read the whole thread):

1. With 26 x 10% equities, you get wiped out if equities drop 38% (assuming no movement in bonds) and experience a margin call even earlier. What are your thoughts on this? Do you expect a flight to quality to offset potential drops in equities? I see you plan/hope for your portfolio to have a max drawdown of 70-80%. Presumably this is because of the expectation of a flight to quality in the event that the stock market tanks?

2. Building on question 1, what about a stagflation scenario? If rates rise and at the same time there is a stock market crash, then you are royally screwed, aren't you? If they happen at the same time, the crash/rise in rates wouldn't even have to be that large, considering your leverage.

3. Have you considered using TIPS to get the flight-to-quality offset against stock market crashes while insulating yourself against the inflation component of stagflation? Of course you'd still be vulnerable to increases in real rates, but it would still be safer, wouldn't it?

4. How exactly did you decide on the specific asset allocation and leverage ratio? By identifying what was, historically, the efficient frontier?

5. What rates are you paying for your short/int/long term treasuries?
Topic Author
Rob Bertram
Posts: 859
Joined: Mon May 05, 2014 12:15 pm

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

Post by Rob Bertram »

Sorry for the slow answer. I've been busy with work and spending time with family over the holidays.
aaaaaaabbbbbbbbbb wrote:Here are a couple questions I have for you Rob (I apologize if they've already been answered, but I don't have time to read the whole thread):

1. With 26 x 10% equities, you get wiped out if equities drop 38% (assuming no movement in bonds) and experience a margin call even earlier. What are your thoughts on this? Do you expect a flight to quality to offset potential drops in equities? I see you plan/hope for your portfolio to have a max drawdown of 70-80%. Presumably this is because of the expectation of a flight to quality in the event that the stock market tanks?
In short, that is half of it. As you point out, the remaining 90% of the portfolio is treasuries leveraged 26x as well. So bonds only need to go up 4% or so to offset the loss in equities. Short-term, I expect a non-trivial amount of flight to quality to offset sharp drops in equities to avoid a "flash crash" style margin call. I am also contributing $3k/month to the portfolio, so that also helps with longer-term bear markets.
aaaaaaabbbbbbbbbb wrote:2. Building on question 1, what about a stagflation scenario? If rates rise and at the same time there is a stock market crash, then you are royally screwed, aren't you? If they happen at the same time, the crash/rise in rates wouldn't even have to be that large, considering your leverage.
Yep, that is possible. I backtested through the 1970s and picked something that would survive through that level of stagflation. It could be worse in the future. In addition, I'm still accumulating, so extra cash will offset some losses and prevent a margin call.

Once I stop adding cash or the portfolio grows large relative to my contributions, I'll reduce leverage.
aaaaaaabbbbbbbbbb wrote:3. Have you considered using TIPS to get the flight-to-quality offset against stock market crashes while insulating yourself against the inflation component of stagflation? Of course you'd still be vulnerable to increases in real rates, but it would still be safer, wouldn't it?
TIPS only protect against unexpected inflation. And there wasn't enough TIPS data for me to feel comfortable using them. Treasuries are what I chose, and I'm happy with that decision.
aaaaaaabbbbbbbbbb wrote:4. How exactly did you decide on the specific asset allocation and leverage ratio? By identifying what was, historically, the efficient frontier?
Some of it was backtesting for an efficient frontier. I also looked at different aspects of the risk profile. For example, maximum drawdown. Global market cap is about 40% stocks and 60% bonds.

Nisi made a good point that bond duration behaves a lot like leverage. He showed that given a stock + long bond portfolio that a stock+ short bond portfolio could be created with the same Sharpe ratio. The 10% stock/ 5% long/85% short treasury portfolio came as a result of applying that to the 40/60 portfolio.
aaaaaaabbbbbbbbbb wrote:5. What rates are you paying for your short/int/long term treasuries?
I'm using futures, so financing for bonds is around the Libor-3month rate.
joglehead
Posts: 9
Joined: Sun Feb 28, 2016 1:55 pm

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

Post by joglehead »

Rob, thanks again for sharing this. I found it tremendously useful. I've been considering starting a similar strategy and have been doing some backtests with portfolio visualizer. It doesn't seem like the long term bonds add much value and in fact they increase the max drawdown at several points throughout history. It seems like an 86/14 ZT/ES portfolio has smaller drawdown and a higher sortino.

I do worry a bit about the stagflation scenario and found that an allocation of 83/13/4 ES/ZT/Gold looks quite good and gives some protection against this. The CAGR and sortino are higher although the max drawdown is also a bit higher as well so you would dial leverage accordingly. I'm curious to hear your thoughts on both of these allocations, thanks!
Topic Author
Rob Bertram
Posts: 859
Joined: Mon May 05, 2014 12:15 pm

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

Post by Rob Bertram »

joglehead wrote:Rob, thanks again for sharing this. I found it tremendously useful. I've been considering starting a similar strategy and have been doing some backtests with portfolio visualizer. It doesn't seem like the long term bonds add much value and in fact they increase the max drawdown at several points throughout history. It seems like an 86/14 ZT/ES portfolio has smaller drawdown and a higher sortino.
For readers who don't have futures symbols memorized:
  • ES = eMini S&P 500 futures contract
  • ZT = 2-Year Treasury note future
Be careful with any futures contract when backtesting using index data. Remember that Portfolio Visualizer does not subtract borrowing costs. There are several periods where borrowing costs exceeded the 2-year note yield. May I suggest using Simba's backtesting spreadsheet. That allows you to enter a negative cash allocation which gives you a better perspective of the risk profile. Also, the spreadsheet breaks the standard deviation between "up" and "down", so you can focus on the left-tail aspect of the risk profile. Also, you will need to add an additional borrowing premium (about 0.3% above the cash/fed funds rate if using futures, 0.5% - 1.5% if using margin).

For example using the spreadsheet, my 10% stock / 5% LT-treasury / 85% ST-treasury x 20 in the spreadsheet would look like this:
200% stocks (using TSM)
100% LT-treasuries
-1900% Tbills/Money market
1700% ST-treasuries
I get a Sortino ratio of 2.01
When I do the same for your 14% stocks / 86% ST-tresuries x20, I get a Sortino ratio of 1.93.
joglehead wrote:I do worry a bit about the stagflation scenario and found that an allocation of 83/13/4 ES/ZT/Gold looks quite good and gives some protection against this. The CAGR and sortino are higher although the max drawdown is also a bit higher as well so you would dial leverage accordingly. I'm curious to hear your thoughts on both of these allocations, thanks!
I will caution anyone modeling gold in the 1970s that it was illegal to invest in gold in the United States from 1933 until the tail end of 1977 (Executive Order 6102). If you start your model in 1978, how does gold look in your portfolio now?
long_gamma
Posts: 420
Joined: Mon Sep 14, 2015 4:13 pm

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

Post by long_gamma »

Have you updated your spreadsheets?

It will be interesting to see recent performance of this strategy. From my reading, it has done extremely well because of heavy leveraged position of bond futures.
"Everyone has a plan 'till they get punched in the mouth." --Mike Tyson
Topic Author
Rob Bertram
Posts: 859
Joined: Mon May 05, 2014 12:15 pm

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

Post by Rob Bertram »

long_gamma wrote:Have you updated your spreadsheets?

It will be interesting to see recent performance of this strategy. From my reading, it has done extremely well because of heavy leveraged position of bond futures.
I haven't updated my spreadsheets since the fall of 2015, but I should. I agree, it would be a great review. As far as leveraged bonds, I can say that futures prices dropped a lot in late 2015 as the Fed raised rates in December and the market expected more increases in 2016. Though, of course, that expectation changed in Jan/Feb, and bond futures prices rebounded. So yes, year-to-date 2016, this strategy did very well. If we look back further, it's not as spectacular but closer to "flat".

For what it's wroth, I've tried to "not peek" as much as I can. I have only logged into my account in 2016 to download tax forms and roll futures contracts.
User avatar
whodidntante
Posts: 13115
Joined: Thu Jan 21, 2016 10:11 pm
Location: outside the echo chamber

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

Post by whodidntante »

I am not an opponent of putting borrowed money into the stock market. I do this myself. However, I suggest you read this before placing a big bet.

http://www.advisorperspectives.com/dsho ... nd-the-SPX
joglehead
Posts: 9
Joined: Sun Feb 28, 2016 1:55 pm

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

Post by joglehead »

Thanks for the feedback Rob. I felt like I was missing something, the projected returns seemed so good it didn't make any sense. I'll try out the spreadsheet you mentioned.

whodidntante, I appreciate your words of caution, but we are talking about leveraging a portfolio that is 90% treasury bonds that will generally perform well when the market goes down. I use margin currently, but futures are new territory for me. Depending on how far I take the leverage, Rob's strategy may actually make my portfolio much less risky and reliant on equity than it is currently.
joglehead
Posts: 9
Joined: Sun Feb 28, 2016 1:55 pm

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

Post by joglehead »

Rob I've been playing with the spreadsheet you mentioned and found that adding some uncorrelated assets appears to have some benefit.

100% Large Cap Blend (most similar to ES right?)
90% LT-treasuries
-1805% Tbills/Money market
1500% ST-treasuries
35% REIT
65% Emerging Mkt
15% Commodities


Sortino if we start in 1978 is 3.008, CAGR 38.47%, max drawdown 85% (1994), 6 total drawdowns with average size of 20%. Am I missing anything?

You would need to use an ETF for REITs and I'm wondering if contango would reduce the long-term performance of commodities when using futures.
User avatar
HomerJ
Posts: 21282
Joined: Fri Jun 06, 2008 12:50 pm

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

Post by HomerJ »

Rob Bertram wrote:Sorry for the slow answer. I've been busy with work and spending time with family over the holidays.
aaaaaaabbbbbbbbbb wrote:Here are a couple questions I have for you Rob (I apologize if they've already been answered, but I don't have time to read the whole thread):

1. With 26 x 10% equities, you get wiped out if equities drop 38% (assuming no movement in bonds) and experience a margin call even earlier. What are your thoughts on this? Do you expect a flight to quality to offset potential drops in equities? I see you plan/hope for your portfolio to have a max drawdown of 70-80%. Presumably this is because of the expectation of a flight to quality in the event that the stock market tanks?
In short, that is half of it. As you point out, the remaining 90% of the portfolio is treasuries leveraged 26x as well. So bonds only need to go up 4% or so to offset the loss in equities.
So, what happens if equities drop and bonds don't go up? or (gasp) they both go down at the same time?

This sounds a lot like those gambling schemes where you make small profits over and over and over until the one low probability (but not zero!) event happens and you get wiped out.
User avatar
HomerJ
Posts: 21282
Joined: Fri Jun 06, 2008 12:50 pm

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

Post by HomerJ »

joglehead wrote:Rob I've been playing with the spreadsheet you mentioned and found that adding some uncorrelated assets appears to have some benefit.
appeared to have some benefit in the PAST. Just be clear about that. You can tweak the model over and over until you find something that works great in back-testing, but that doesn't mean it will work going forward.
lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 10:55 pm

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

Post by lack_ey »

joglehead wrote:Rob I've been playing with the spreadsheet you mentioned and found that adding some uncorrelated assets appears to have some benefit.

100% Large Cap Blend (most similar to ES right?)
90% LT-treasuries
-1805% Tbills/Money market
1500% ST-treasuries
35% REIT
65% Emerging Mkt
15% Commodities


Sortino if we start in 1978 is 3.008, CAGR 38.47%, max drawdown 85% (1994), 6 total drawdowns with average size of 20%. Am I missing anything?

You would need to use an ETF for REITs and I'm wondering if contango would reduce the long-term performance of commodities when using futures.
Wait, which spreadsheet? That would determine the data source for commodities, which might already be futures and not spot price. It depends.

And I guess it's using cash rate for borrowing?

Yeah, if you can actually borrow at the risk-free rate all the times, you could add extra stuff easily and probably have a better chance of coming out better. If you pay a premium over the risk-free rate as with futures and other derivatives (never mind actual margin for retail investors, which is worse), it doesn't work as well in practice.
User avatar
rmelvey
Posts: 826
Joined: Sat Sep 18, 2010 5:17 pm
Contact:

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

Post by rmelvey »

What are the thoughts on the tax efficiency of this? Do you have money in retirement accounts? Are those seen as part of this strategy or are they not factored into this asset allocation?

I like that rolling futures of a bond position allows you to book part of it as capital gains, but better yet is holding regular bond funds in a tax deferred account.
joglehead
Posts: 9
Joined: Sun Feb 28, 2016 1:55 pm

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

Post by joglehead »

lack_ey, the spreadsheet Rob pointed me towards is here.

The data sources listed are below. I'm not sure how they are used collectively in the spreadsheet.
COMM - Commodities/Natural Resources
Collaterallized Chase Index 1972-1990
DJ-AIJ plus T-Bills 1991-1996
DJ-AIG plus Yahoo's IPS 'category' returns 1997-2001 (prior to VIPSX)
DJ-AIJ plus VIPSX - 2001-2002
Pimco PCRIX 2003+

This is assuming the T-bill rate for borrowing. There are a lot of asset classes you can add that actually reduce the sortino ratio, but is there a better way we can approximate borrowing costs?

I agree we have to be careful with backtesting and there is no free lunch, any portfolio has risk and we don't know what the future holds. I think this makes it more prudent to add different types of assets that can hold up well in stagflation or when bonds/stocks both go down.

I did notice something strange when backtesting to 1871. The rate for the T-bill seems to be higher than the 2-year for much of 1871-1930. Almost double in fact for much of it. Anyone have insight as to why? I think this highlights the need to dynamically shift out of this strategy as Rob mentions if this portion of the curve inverts.
Topic Author
Rob Bertram
Posts: 859
Joined: Mon May 05, 2014 12:15 pm

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

Post by Rob Bertram »

joglehead wrote:Rob I've been playing with the spreadsheet you mentioned and found that adding some uncorrelated assets appears to have some benefit.

100% Large Cap Blend (most similar to ES right?)
90% LT-treasuries
-1805% Tbills/Money market
1500% ST-treasuries
35% REIT
65% Emerging Mkt
15% Commodities


Sortino if we start in 1978 is 3.008, CAGR 38.47%, max drawdown 85% (1994), 6 total drawdowns with average size of 20%. Am I missing anything?

You would need to use an ETF for REITs and I'm wondering if contango would reduce the long-term performance of commodities when using futures.
This looks a little like a leveraged version of the Larry portfolio. My main concern with a portfolio that overweights EM is that it's based on some one-time events. I'm not a big fan of commodities, either. There is no expectation that commodities will be worth more tomorrow than today. They don't pay dividends, etc.

My original core portfolio was built partly on backtesting but also on market cap of stocks and bonds. When we were discussing back in 2014, bonds had a global market capitalization of about $83 trillion, and stocks were around $55 trillion if I remember correctly. So a total market portfolio of stocks and bonds would be about 40% stocks and 60% bonds. I haven't checked to see how that has changed in two years. I changed from 40/60 to include short-term bonds based on a thread by Nisiprius where he argued that bond duration can be explained by leverage. He showed that a similar portfolio of stocks and short-term bonds could be created to match the Sharpe ratio of a stock and intermediate-term (or long-term) bond portfolio.

If I could get the same margin requirements and financing rates with a regular margin account as I can with futures, I would just leverage total-market ETFs. Instead, I use the S&P 500 as a proxy for the total market and treasuries as a proxy for bonds. If I get to $5 million in margin, IB's margin rates would start to become competitive with futures financing rates. Plus, there would be a mental benefit of switching from futures to ETFs in that I wouldn't need to worry about rolling anymore.
Topic Author
Rob Bertram
Posts: 859
Joined: Mon May 05, 2014 12:15 pm

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

Post by Rob Bertram »

rmelvey wrote:What are the thoughts on the tax efficiency of this? Do you have money in retirement accounts? Are those seen as part of this strategy or are they not factored into this asset allocation?

I like that rolling futures of a bond position allows you to book part of it as capital gains, but better yet is holding regular bond funds in a tax deferred account.
We had some discussions about tax efficiency of this methodology several pages back. Ideally, keep as much of the stock portion in ETFs so that cap gains can be held (deferred) long term. Bond futures are a little more tax efficient due to the 60/40% long-term/short-term tax treatment.

I do have this in a taxable account. My retirement savings are separate and are invested in a sane 3-fund portfolio. All tax-advantaged accounts for me and my wife are maxed out every year, so I'm not jeopardizing our retirement in any way.

Assuming that my leverage experiment works successfully over the next 5 years or so, I might consider doing this in a Roth account as well. Though, I will definitely discuss it here before I do it. I can't borrow in a retirement account, but a few brokers allow futures trading (ThinkOrSwim and Interactive Brokers).
Park
Posts: 738
Joined: Sat Nov 06, 2010 4:56 pm

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

Post by Park »

There should be a warning put on this thread to not do this at home.

I'm not questioning whether the ideas and their implementation are wrong. But for many who read this thread, such as myself ;-), this is above their level of knowledge.

I'm concerned that someone or some people will hurt themselves financially by trying to imitate what's being done here.
Topic Author
Rob Bertram
Posts: 859
Joined: Mon May 05, 2014 12:15 pm

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

Post by Rob Bertram »

I think I'm fairly clear in the OP as to the constraints of this experiment in terms of maxing tax advantaged accounts first, etc.

Besides saying that people need to assess their own need, ability, and willing to take risk, what more can I do to caution people?
Aish
Posts: 128
Joined: Thu Mar 20, 2014 7:49 pm

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

Post by Aish »

I think another way to maybe construct this, but without necessarily using margin for bonds is potentially buying LEAPs deep ITM (delta=1) on the S&P 500. They're priced at very close to 0 time value when you consider interest rates and dividends. Also, it has the benefit of not forcing a margin call. Of course, given certain strike prices you could lose the entire value of the option, but a) you can choose a lower strike price (and less return obviously), b) you don't face margin calls, so it's still not as bad. Something worth thinking about for people interested in leverage.
Topic Author
Rob Bertram
Posts: 859
Joined: Mon May 05, 2014 12:15 pm

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

Post by Rob Bertram »

Nalebuff and Ayres give special mention to LEAPs in their Lifecycle Investing book. And if I remember correctly, they have been discussed briefly in this thread. I believe the major problem is liquidity, and exercising early exacerbates the liquidity problem.

search.php?keywords=LEAPs&t=143037&sf=msgonly
joglehead
Posts: 9
Joined: Sun Feb 28, 2016 1:55 pm

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

Post by joglehead »

Hi Rob, I've again been reviewing portfolio visualizer and the backtest spreadsheet and it seems like 5 year treasury bonds have better risk-adjusted returns than 2 year treasury bonds. Do you get something different?
Day9
Posts: 1000
Joined: Mon Jun 11, 2012 6:22 pm

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

Post by Day9 »

joglehead wrote:Hi Rob, I've again been reviewing portfolio visualizer and the backtest spreadsheet and it seems like 5 year treasury bonds have better risk-adjusted returns than 2 year treasury bonds. Do you get something different?
I've found that too. 5 years has been something of a "sweet spot" in the past. But with this strategy it is more important than ever to look at an investment as part of a portfolio instead of in isolation. I believe Rob already posted his findings of historical risk and max drawdown of the different portfolios using different treasury durations and leverage.

A good exercise would be to compare the historical correlation of different treasury durations with stocks, and how they behaved during financial crises, stock market crashes, and prolonged bear markets. Real returns are important, of course, but Nominal returns might also be independently important since they are related to margin call.
I'm just a fan of the person I got my user name from
Topic Author
Rob Bertram
Posts: 859
Joined: Mon May 05, 2014 12:15 pm

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

Post by Rob Bertram »

joglehead wrote:Hi Rob, I've again been reviewing portfolio visualizer and the backtest spreadsheet and it seems like 5 year treasury bonds have better risk-adjusted returns than 2 year treasury bonds. Do you get something different?
Expected risk-adjusted returns (Sharpe Ratio) is definitely something to consider, but as Day9 points out max drawdown is more important. Here's an example:
Image
Portfolio 1 is 40% stocks / 60% 5-year treasuries
Portfolio 2 is 10% stocks / 5% long-term treasuries / 85% 2-year treasures.
Portfolio 3 is 100% cash

The Sharpe ratios are roughly the same (portfolio 1 being slightly higher). After subtracting the cost of borrowing, the max drawdown is
Portfolio 1: 17% max drawdown
Portfolio 2: 6% max drawdown.

Now, let's say that my risk tolerance is 50% loss. With portfolio 1, that means I can do about 50/17 = 2.94x leverage. With Portfolio 2, I can go 50/6 = 8.33x leverage. That gives me an expected return for portfolio 1: (8.88-4.92)*2.94 = 11.64%. And for portfolio 2: (6.84-4.92)*8.33 = 15.99%
Post Reply