Are 3x leveraged ETFs the long-term winning strategy?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
EfficientInvestor
Posts: 421
Joined: Thu Nov 01, 2018 7:02 pm
Location: Alabama

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by EfficientInvestor »

tradri wrote: Mon Apr 12, 2021 9:45 am
EfficientInvestor wrote: Mon Apr 12, 2021 9:30 am Why do you feel the need to reset the leverage so often? How often do your reset allocations on your existing portfolios? Why would you want to do the same thing that the Leveraged ETFs do under the hood? The daily reset of leverage is what leads to the volatility decay. I would say that the lack of rebalancing of a futures portfolio is a benefit to using futures. To help answer the question, I'll point to this backtest: https://www.portfoliovisualizer.com/bac ... on5_2=-200

It is 50/50 UPRO/TMF vs 150/150 SPY/TLT. The SPY/TLT portfolio represents the use of futures contracts to implement the strategy. It is rebalanced once per year. If I remove the LETFs from the backtest and go further back in time using VFINX/VUSTX instead of SPY/TLT, we get this: https://www.portfoliovisualizer.com/bac ... on5_2=-200

The point I'm trying to make with both backtests is that the portfolio has done just fine with only rebalancing once per year. As always, this backtest is no guarantee of the future. The futures-based portfolio is definitely prone to wild swings in overall exposure to the various assets in the portfolio, but isn't that the case with all investing? Just make sure your investment policy statement has rules for when to rebalance the same way you would a non-leveraged portfolio.
Interesting.

So you are basically saying that if you 1.5x the stock market and 1.5x the bond market, there is practically no risk of losing all your money, unless the stock/bond market drop by ~66%? And if it comes close to that, you can just rebalance to buy the dip?

Somehow this seems too good to be true. 20.19% CAGR over 40 years with a max drawdown of only -56%?? What's the catch?
The catch is that this backtest is limited to a period when bonds were in a bull market. The 40 year period prior to it wasn’t so attractive due to the rise in interest rates.

Edit/Addition - from 1955-1981, the 150/150 portfolio would have had around a 3% return per year and a max drawdown around -80%.
User avatar
Topic Author
tradri
Posts: 549
Joined: Tue Mar 23, 2021 5:42 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by tradri »

EfficientInvestor wrote: Mon Apr 12, 2021 9:56 am
The catch is that this backtest is limited to a period when bonds were in a bull market. The 40 year period prior to it wasn’t so attractive due to the rise in interest rates.
But even then, you could just own an uncorrelated asset like gold.

I am more concerned about the practical implementation of such a strategy. What are the benefits of doing that versus just getting a loan/buying on margin? Aren't there also costs when buying/selling options yourself?
EfficientInvestor
Posts: 421
Joined: Thu Nov 01, 2018 7:02 pm
Location: Alabama

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by EfficientInvestor »

tradri wrote: Mon Apr 12, 2021 10:02 am
EfficientInvestor wrote: Mon Apr 12, 2021 9:56 am
The catch is that this backtest is limited to a period when bonds were in a bull market. The 40 year period prior to it wasn’t so attractive due to the rise in interest rates.
But even then, you could just own an uncorrelated asset like gold.

I am more concerned about the practical implementation of such a strategy. What are the benefits of doing that versus just getting a loan/buying on margin? Aren't there also costs when buying/selling options yourself?
I agree that the addition of more uncorrelated assets helps mitigate the risks. I also include gold in my portfolio.

As for practical implementation, the benefit of using futures over loan/margin is borrowing cost. When using futures, you are essentially borrowing at the risk-free rate. With margin, you are borrowing at the risk-free rate plus the interest rate your broker charges.
User avatar
Topic Author
tradri
Posts: 549
Joined: Tue Mar 23, 2021 5:42 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by tradri »

EfficientInvestor wrote: Mon Apr 12, 2021 10:11 am
I agree that the addition of more uncorrelated assets helps mitigate the risks. I also include gold in my portfolio.

As for practical implementation, the benefit of using futures over loan/margin is borrowing cost. When using futures, you are essentially borrowing at the risk-free rate. With margin, you are borrowing at the risk-free rate plus the interest rate your broker charges.
And how exactly do you handle the change in your leverage ratio? You just buy options/futures with 1.5x leverage (Omega?) to the underlying index and sell them after one year to replace them with new ones?

Do you also account for crazy movements? Say if the underlying drops by -50%, do you sell the option and replace it with a new one?
User avatar
Topic Author
tradri
Posts: 549
Joined: Tue Mar 23, 2021 5:42 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by tradri »

EfficientInvestor wrote: Mon Apr 12, 2021 10:11 am
I agree that the addition of more uncorrelated assets helps mitigate the risks. I also include gold in my portfolio.

As for practical implementation, the benefit of using futures over loan/margin is borrowing cost. When using futures, you are essentially borrowing at the risk-free rate. With margin, you are borrowing at the risk-free rate plus the interest rate your broker charges.
Also, what are the practical difference of using futures vs options? The only thing I hear is "the right but not the obligation to buy/sell". But which one is better for achieving a leveraged position? Is it also true that futures can only be bought in big packages, like $100k?
EfficientInvestor
Posts: 421
Joined: Thu Nov 01, 2018 7:02 pm
Location: Alabama

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by EfficientInvestor »

tradri wrote: Mon Apr 12, 2021 10:16 am
EfficientInvestor wrote: Mon Apr 12, 2021 10:11 am
I agree that the addition of more uncorrelated assets helps mitigate the risks. I also include gold in my portfolio.

As for practical implementation, the benefit of using futures over loan/margin is borrowing cost. When using futures, you are essentially borrowing at the risk-free rate. With margin, you are borrowing at the risk-free rate plus the interest rate your broker charges.
And how exactly do you handle the change in your leverage ratio? You just buy options/futures with 1.5x leverage (Omega?) to the underlying index and sell them after one year to replace them with new ones?

Do you also account for crazy movements? Say if the underlying drops by -50%, do you sell the option and replace it with a new one?
tradri wrote: Mon Apr 12, 2021 10:37 am Also, what are the practical difference of using futures vs options? The only thing I hear is "the right but not the obligation to buy/sell". But which one is better for achieving a leveraged position? Is it also true that futures can only be bought in big packages, like $100k?
I don't really look to handle the changes in the leverage ratio. That's not to say I'm not aware of them. I just accept them for what they are. I'm especially hesitant to do any rebalancing more than once per year if I'm in a taxable account due to the difference in short term and long term tax rates. However, due to the tax inefficiency of derivatives, I usually look to use them in tax-advantaged accounts anyway. If I'm in a tax-advantaged account, I don't see much value in rebalancing more than once per year. I haven't seen enough significant evidence to support rebalancing more often than annually. However, if you are using futures contracts and have to roll them once per quarter anyway, you might as well take the opportunity to do a rebalance as well.

As for futures vs options...I would say futures are just pure leverage. That is why you can approximate their historical performance in PortfolioVisualizer using negative CASHX. On the other hand, options are leverage and insurance all wrapped into one. If you just want the leverage, it will be cheaper to use futures because they don't incorporate payment for the insurance in the price of the contract in the way that options do. If you are looking to do something as leveraged as a 150/150 portfolio, it might be worthwhile to consider using options to protect against outlier events. But the use of options for the purpose of leveraging and protecting a portfolio is a whole other ball of wax that merits its own thread, so I'll leave it at that for now.
User avatar
Topic Author
tradri
Posts: 549
Joined: Tue Mar 23, 2021 5:42 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by tradri »

EfficientInvestor wrote: Mon Apr 12, 2021 12:19 pm I don't really look to handle the changes in the leverage ratio. That's not to say I'm not aware of them. I just accept them for what they are. I'm especially hesitant to do any rebalancing more than once per year if I'm in a taxable account due to the difference in short term and long term tax rates. However, due to the tax inefficiency of derivatives, I usually look to use them in tax-advantaged accounts anyway. If I'm in a tax-advantaged account, I don't see much value in rebalancing more than once per year. I haven't seen enough significant evidence to support rebalancing more often than annually. However, if you are using futures contracts and have to roll them once per quarter anyway, you might as well take the opportunity to do a rebalance as well.

As for futures vs options...I would say futures are just pure leverage. That is why you can approximate their historical performance in PortfolioVisualizer using negative CASHX. On the other hand, options are leverage and insurance all wrapped into one. If you just want the leverage, it will be cheaper to use futures because they don't incorporate payment for the insurance in the price of the contract in the way that options do. If you are looking to do something as leveraged as a 150/150 portfolio, it might be worthwhile to consider using options to protect against outlier events. But the use of options for the purpose of leveraging and protecting a portfolio is a whole other ball of wax that merits its own thread, so I'll leave it at that for now.
Honestly, when looking at a quarterly rebalanced strategy for 70/30 UPRO/TMF vs 210/90/-200 SPY/TLT/CASHX, the difference in CAGR between those 2 strategies isn't all that far apart. (since the inception of these funds) https://www.portfoliovisualizer.com/bac ... tion5_2=30

The 210/90/-200 SPY/TLT/CASHX futures strategy does produce a ~2% higher CAGR over that time period, but aren't there a lot of other issues when using futures? In the original HFEA thread, he said that futures can only be bought for $140k a piece, which would make it very difficult to get started & later on size your positions. Also, isn't it possible to lose more than your initial investment with futures, since you have the right and obligation to fulfill the future at the end? Doesn't this also mean that one has to have substantial collateral in the account, and therefore the risk of getting margin called?

And for options, as you said, the CAGR will be lower than the -CASHX Portfolio Visualizer approximation, so they don't really provide an advantage over leveraged ETFs, right?
EfficientInvestor
Posts: 421
Joined: Thu Nov 01, 2018 7:02 pm
Location: Alabama

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by EfficientInvestor »

tradri wrote: Mon Apr 12, 2021 1:01 pm Honestly, when looking at a quarterly rebalanced strategy for 70/30 UPRO/TMF vs 210/90/-200 SPY/TLT/CASHX, the difference in CAGR between those 2 strategies isn't all that far apart. (since the inception of these funds) https://www.portfoliovisualizer.com/bac ... tion5_2=30

The 210/90/-200 SPY/TLT/CASHX futures strategy does produce a ~2% higher CAGR over that time period, but aren't there a lot of other issues when using futures? In the original HFEA thread, he said that futures can only be bought for $140k a piece, which would make it very difficult to get started & later on size your positions. Also, isn't it possible to lose more than your initial investment with futures, since you have the right and obligation to fulfill the future at the end? Doesn't this also mean that one has to have substantial collateral in the account, and therefore the risk of getting margin called?

And for options, as you said, the CAGR will be lower than the -CASHX Portfolio Visualizer approximation, so they don't really provide an advantage over leveraged ETFs, right?
The monthly rebal and annual rebal both have a CAGR difference of around 5%, not just 2%. I would say the 2% difference is just the result of good rebal timing that you can't necessarily expect going forward.

The e-mini micro contracts currently go for around $20k, so that isn't too bad. The size of the treasury contracts are definitely limiting since they are $100k or more.

You can definitely lose more than your initial investment with futures and this needs to be accounted for in how you manage the portfolio. I keep a certain amount of cash on hand to handle the daily fluctuations. I keep any additional unused capital in an ultra-short term bond fund like JPST. I would never do a 210/90 portfolio with just futures. You either need to lower the leverage or use options to cover the worst case scenarios so you don't lose all your money. For someone that otherwise wants similar volatility to 100% stock, you would probably want something more like 90% stock and 60% LTT.

Regarding options...you can't really do a comparison using Portfolio Visualizer. The cost and benefit of the options can't be reflected in the backtest. I personally buy put options (or in the money call options) 2 years out in time to serve as my downside protection and then offset the price of those by selling monthly covered calls against 25-33% of my holdings. This keeps the portfolio around theta neutral so that the daily premium received from the covered calls will theoretically offset the daily premium paid out for the protection. The intent of doing all of this is to try to hedge black swan events by giving up a little of your potential upside. By applying the hedges, you are giving up some potential return, but you are also reducing standard deviation. Overall, the goal is to end up with higher risk-adjusted returns by mitigating the really large drawdown event.
User avatar
Topic Author
tradri
Posts: 549
Joined: Tue Mar 23, 2021 5:42 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by tradri »

EfficientInvestor wrote: Mon Apr 12, 2021 1:46 pm The monthly rebal and annual rebal both have a CAGR difference of around 5%, not just 2%. I would say the 2% difference is just the result of good rebal timing that you can't necessarily expect going forward.

The e-mini micro contracts currently go for around $20k, so that isn't too bad. The size of the treasury contracts are definitely limiting since they are $100k or more.

You can definitely lose more than your initial investment with futures and this needs to be accounted for in how you manage the portfolio. I keep a certain amount of cash on hand to handle the daily fluctuations. I keep any additional unused capital in an ultra-short term bond fund like JPST. I would never do a 210/90 portfolio with just futures. You either need to lower the leverage or use options to cover the worst case scenarios so you don't lose all your money. For someone that otherwise wants similar volatility to 100% stock, you would probably want something more like 90% stock and 60% LTT.

Regarding options...you can't really do a comparison using Portfolio Visualizer. The cost and benefit of the options can't be reflected in the backtest. I personally buy put options (or in the money call options) 2 years out in time to serve as my downside protection and then offset the price of those by selling monthly covered calls against 25-33% of my holdings. This keeps the portfolio around theta neutral so that the daily premium received from the covered calls will theoretically offset the daily premium paid out for the protection. The intent of doing all of this is to try to hedge black swan events by giving up a little of your potential upside. By applying the hedges, you are giving up some potential return, but you are also reducing standard deviation. Overall, the goal is to end up with higher risk-adjusted returns by mitigating the really large drawdown event.
All these advanced "Hedgefund" strategies of leveraging this, shorting that, hedging black swans, etc... seem a little bit unnecessary to me if I am just looking for a portfolio that will do well long-term when leveraged.

In the beginning you mention that the Portfolio Visualizer -CASHX futures portfolio outperforms the leveraged ETFs, but later on you say that the risk of losing money is too great, and that one should lower the leverage or hedge the risk. Doesn't this "hedging", then result in lower returns than the pure leveraged ETF portfolio?

Just to be clear, I am not interested in using some advanced hedging strategies to improve my risk/reward ratio. I am just interest in the highest CAGR long-term.
EfficientInvestor
Posts: 421
Joined: Thu Nov 01, 2018 7:02 pm
Location: Alabama

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by EfficientInvestor »

tradri wrote: Mon Apr 12, 2021 2:04 pm All these advanced "Hedgefund" strategies of leveraging this, shorting that, hedging black swans, etc... seem a little bit unnecessary to me if I am just looking for a portfolio that will do well long-term when leveraged.

In the beginning you mention that the Portfolio Visualizer -CASHX futures portfolio outperforms the leveraged ETFs, but later on you say that the risk of losing money is too great, and that one should lower the leverage or hedge the risk. Doesn't this "hedging", then result in lower returns than the pure leveraged ETF portfolio?

Just to be clear, I am not interested in using some advanced hedging strategies to improve my risk/reward ratio. I am just interest in the highest CAGR long-term.
Well what I'm saying is that once your leverage gets beyond a certain point, you have to start thinking about total loss events. One way to mitigate the total loss event is to just use LETFs and be done with it. This sounds like the best route for you. The other way to mitigate the risk of the total loss event is to hedge the portfolio with options. Hence the reason for going down the road of explaining all of that.
User avatar
Topic Author
tradri
Posts: 549
Joined: Tue Mar 23, 2021 5:42 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by tradri »

EfficientInvestor wrote: Mon Apr 12, 2021 2:09 pm Well what I'm saying is that once your leverage gets beyond a certain point, you have to start thinking about total loss events. One way to mitigate the total loss event is to just use LETFs and be done with it. This sounds like the best route for you. The other way to mitigate the risk of the total loss event is to hedge the portfolio with options. Hence the reason for going down the road of explaining all of that.
I see. And you believe (know) that such a directly leveraged futures portfolio hedged with options will produce a higher CAGR than, for example, a 70/30 UPRO/TMF portfolio rebalanced quarterly?
EfficientInvestor
Posts: 421
Joined: Thu Nov 01, 2018 7:02 pm
Location: Alabama

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by EfficientInvestor »

tradri wrote: Mon Apr 12, 2021 2:12 pm
EfficientInvestor wrote: Mon Apr 12, 2021 2:09 pm Well what I'm saying is that once your leverage gets beyond a certain point, you have to start thinking about total loss events. One way to mitigate the total loss event is to just use LETFs and be done with it. This sounds like the best route for you. The other way to mitigate the risk of the total loss event is to hedge the portfolio with options. Hence the reason for going down the road of explaining all of that.
I see. And you believe (know) that such a directly leveraged futures portfolio hedged with options will produce a higher CAGR than, for example, a 70/30 UPRO/TMF portfolio rebalanced quarterly?
I don't know it, but I believe it.
User avatar
Topic Author
tradri
Posts: 549
Joined: Tue Mar 23, 2021 5:42 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by tradri »

EfficientInvestor wrote: Mon Apr 12, 2021 2:18 pm I don't know it, but I believe it.
How do you make sure that the strategy has worked in the past/will work in the future, if you said that it's very difficult to backtest these options?
DMoogle
Posts: 171
Joined: Sat Oct 31, 2020 10:24 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by DMoogle »

tradri wrote: Mon Apr 12, 2021 2:26 pm How do you make sure that the strategy has worked in the past/will work in the future, if you said that it's very difficult to backtest these options?
Nobody knows, homie. The theory is sound though. A DIY method is fundamentally similar to what some LETFs do themselves, only without the expense ratio and daily rebalancing.

Removing daily rebalancing is a double-edged sword because, while it reduces volatility drag, it also means your leverage ratio won't stay constant. Futures and options aren't some exotic Pandora's box... their values are just based on the values of the underlying financial instruments.
User avatar
Topic Author
tradri
Posts: 549
Joined: Tue Mar 23, 2021 5:42 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by tradri »

DMoogle wrote: Mon Apr 12, 2021 2:36 pm Nobody knows, homie. The theory is sound though. A DIY method is fundamentally similar to what some LETFs do themselves, only without the expense ratio and daily rebalancing.

Removing daily rebalancing is a double-edged sword because, while it reduces volatility drag, it also means your leverage ratio won't stay constant. Futures and options aren't some exotic Pandora's box... their values are just based on the values of the underlying financial instruments.
I find it hard to believe that you will incur fewer costs when trading these futures/options by yourself, than the 0.93% annual expense of these leveraged ETFs.

As far as the daily rebalancing is concerned, I agree that it is a double-edged sword. The problem is, that the longer you extend the time period between the rebalancings, the more likely you are to face total loss (given enough leverage). The certainty, that the stock market can't drop by more than 20% a day due to circuit breakers, is what made these daily leveraged ETFs so appealing to me in the first place.

As far as hedging the downside with options, I have no idea what the risk/reward characteristics of that could be...
EfficientInvestor
Posts: 421
Joined: Thu Nov 01, 2018 7:02 pm
Location: Alabama

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by EfficientInvestor »

tradri wrote: Mon Apr 12, 2021 2:26 pm
EfficientInvestor wrote: Mon Apr 12, 2021 2:18 pm I don't know it, but I believe it.
How do you make sure that the strategy has worked in the past/will work in the future, if you said that it's very difficult to backtest these options?
My backtesting is still built around the underlying allocations. But then, on top of that, I take into account the contractual upside and downside of the option contracts in order to determine how my portfolio would act at the extremes. For instance, if I buy put options against all holdings, I know what my max drawdown would be because it is contractually limited. If the price of the underlying goes through the roof and I'm selling covered calls against 25% of holdings, I know that my upside will be capped on that portion.

I have used TD Ameritrade's Thinkorswim platform to backtest a few options strategies, but you start to run into data limitations the further you go back in time. I find that doing the more high level approach detailed above is sufficient for my purposes.
User avatar
Topic Author
tradri
Posts: 549
Joined: Tue Mar 23, 2021 5:42 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by tradri »

EfficientInvestor wrote: Mon Apr 12, 2021 2:51 pm My backtesting is still built around the underlying allocations. But then, on top of that, I take into account the contractual upside and downside of the option contracts in order to determine how my portfolio would act at the extremes. For instance, if I buy put options against all holdings, I know what my max drawdown would be because it is contractually limited. If the price of the underlying goes through the roof and I'm selling covered calls against 25% of holdings, I know that my upside will be capped on that portion.

I have used TD Ameritrade's Thinkorswim platform to backtest a few options strategies, but you start to run into data limitations the further you go back in time. I find that doing the more high level approach detailed above is sufficient for my purposes.
And how has this strategy performed until now (or based on the backtest)?

Has it beaten a 70/30 UPRO/TMF portfolio rebalanced quarterly over that time period? If yes, by how much?
DMoogle
Posts: 171
Joined: Sat Oct 31, 2020 10:24 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by DMoogle »

tradri wrote: Mon Apr 12, 2021 2:46 pmI find it hard to believe that you will incur fewer costs when trading these futures/options by yourself, than the 0.93% annual expense of these leveraged ETFs.
Not sure about that one... Interactive Brokers charges $0.25 per contract, which is nothing if you're rebalancing annually or even quarterly. While you *might* might pay slightly more in spreads than a large institution like ProShares, the cost of trading is de minimis. S&P500 and treasury futures and options are commonly traded instruments as well, so it's possible you'll be paying even lower spreads than ProShares does on its swaps.

0.93% annual expense just goes right into ProShare's pocket. The cost of spreads and commission fees, as you know, still have to be paid out of the fund. I think of it as a convenience fee.
User avatar
Topic Author
tradri
Posts: 549
Joined: Tue Mar 23, 2021 5:42 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by tradri »

DMoogle wrote: Mon Apr 12, 2021 3:04 pm I think of it as a convenience fee.
Yeah, that could very well be true. (assuming quarterly or annual re-leveraging is more ideal)
alex_686
Posts: 8303
Joined: Mon Feb 09, 2015 2:39 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by alex_686 »

EfficientInvestor wrote: Mon Apr 12, 2021 1:46 pm The monthly rebal and annual rebal both have a CAGR difference of around 5%, not just 2%. I would say the 2% difference is just the result of good rebal timing that you can't necessarily expect going forward.
It is more subtle than that. It is very easy to simulate situations where the values would be reversed.

As I have said before, it depends on volatility, returns, and if the market is trend following or mean reverting. If there is high volatility and mean reversion then frequent rebalancing will generate higher returns. If not, lower returns. Different periods will get you different results.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
EfficientInvestor
Posts: 421
Joined: Thu Nov 01, 2018 7:02 pm
Location: Alabama

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by EfficientInvestor »

alex_686 wrote: Mon Apr 12, 2021 3:09 pm
EfficientInvestor wrote: Mon Apr 12, 2021 1:46 pm The monthly rebal and annual rebal both have a CAGR difference of around 5%, not just 2%. I would say the 2% difference is just the result of good rebal timing that you can't necessarily expect going forward.
It is more subtle than that. It is very easy to simulate situations where the values would be reversed.

As I have said before, it depends on volatility, returns, and if the market is trend following or mean reverting. If there is high volatility and mean reversion then frequent rebalancing will generate higher returns. If not, lower returns. Different periods will get you different results.
I agree. I guess the point I'm trying to make is that there is no way to really know what the optimal rebal strategy will be going forward. Attempting to optimize a rebal strategy of LETFs that beats the use of futures contracts would take much more effort (if even possible), than just implementing with futures.
User avatar
RovenSkyfall
Posts: 245
Joined: Wed Apr 01, 2020 11:40 am

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by RovenSkyfall »

tradri wrote: Mon Apr 12, 2021 2:46 pm
DMoogle wrote: Mon Apr 12, 2021 2:36 pm Nobody knows, homie. The theory is sound though. A DIY method is fundamentally similar to what some LETFs do themselves, only without the expense ratio and daily rebalancing.

Removing daily rebalancing is a double-edged sword because, while it reduces volatility drag, it also means your leverage ratio won't stay constant. Futures and options aren't some exotic Pandora's box... their values are just based on the values of the underlying financial instruments.
I find it hard to believe that you will incur fewer costs when trading these futures/options by yourself, than the 0.93% annual expense of these leveraged ETFs.

As far as the daily rebalancing is concerned, I agree that it is a double-edged sword. The problem is, that the longer you extend the time period between the rebalancings, the more likely you are to face total loss (given enough leverage). The certainty, that the stock market can't drop by more than 20% a day due to circuit breakers, is what made these daily leveraged ETFs so appealing to me in the first place.

As far as hedging the downside with options, I have no idea what the risk/reward characteristics of that could be...
I believe Uncorrelated did the math in a post somewhere and found that if you want more than 2x leverage LETFs were a better option. If you want somewhere between 1-2x leverage then futures were cheaper.
I saved my money, but it can't save me | The Chariot
User avatar
Topic Author
tradri
Posts: 549
Joined: Tue Mar 23, 2021 5:42 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by tradri »

EfficientInvestor wrote: Mon Apr 12, 2021 3:13 pm I agree. I guess the point I'm trying to make is that there is no way to really know what the optimal rebal strategy will be going forward. Attempting to optimize a rebal strategy of LETFs that beats the use of futures contracts would take much more effort (if even possible), than just implementing with futures.
I am more interested in the cost/opportunity cost of the money you said was used to either hedge the downside risk with options or sits in cash/very short term bonds. Doesn't that limit the upside you get of using futures?
Last edited by tradri on Mon Apr 12, 2021 3:32 pm, edited 1 time in total.
User avatar
Topic Author
tradri
Posts: 549
Joined: Tue Mar 23, 2021 5:42 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by tradri »

RovenSkyfall wrote: Mon Apr 12, 2021 3:19 pm
I believe Uncorrelated did the math in a post somewhere and found that if you want more than 2x leverage LETFs were a better option. If you want somewhere between 1-2x leverage then futures were cheaper.
Could you link that post, please?
EfficientInvestor
Posts: 421
Joined: Thu Nov 01, 2018 7:02 pm
Location: Alabama

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by EfficientInvestor »

tradri wrote: Mon Apr 12, 2021 3:00 pm
EfficientInvestor wrote: Mon Apr 12, 2021 2:51 pm My backtesting is still built around the underlying allocations. But then, on top of that, I take into account the contractual upside and downside of the option contracts in order to determine how my portfolio would act at the extremes. For instance, if I buy put options against all holdings, I know what my max drawdown would be because it is contractually limited. If the price of the underlying goes through the roof and I'm selling covered calls against 25% of holdings, I know that my upside will be capped on that portion.

I have used TD Ameritrade's Thinkorswim platform to backtest a few options strategies, but you start to run into data limitations the further you go back in time. I find that doing the more high level approach detailed above is sufficient for my purposes.
And how has this strategy performed until now (or based on the backtest)?

Has it beaten a 70/30 UPRO/TMF portfolio rebalanced quarterly over that time period? If yes, by how much?
The strategy has performed well since I started in ~2017. However, it is somewhat meaningless to compare it to 70/30 UPRO/TMF because it is not apples to apples. My portfolio is much more diversified than that and any differences in performance would be just as much due to differences in underlying allocation as it would be due to the method of implementation (which is the question currently at hand).
User avatar
Topic Author
tradri
Posts: 549
Joined: Tue Mar 23, 2021 5:42 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by tradri »

EfficientInvestor wrote: Mon Apr 12, 2021 3:42 pm The strategy has performed well since I started in ~2017. However, it is somewhat meaningless to compare it to 70/30 UPRO/TMF because it is not apples to apples. My portfolio is much more diversified than that and any differences in performance would be just as much due to differences in underlying allocation as it would be due to the method of implementation (which is the question currently at hand).
I agree, comparing it to that wouldn't be fair in that short time period.

Maybe it's more fair to compare your portfolio to a 3x leveraged ETF portfolio representing your asset allocation?
Maybe to something like UPRO/TMF/UGLD/etc...?
EfficientInvestor
Posts: 421
Joined: Thu Nov 01, 2018 7:02 pm
Location: Alabama

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by EfficientInvestor »

tradri wrote: Mon Apr 12, 2021 3:45 pm
EfficientInvestor wrote: Mon Apr 12, 2021 3:42 pm The strategy has performed well since I started in ~2017. However, it is somewhat meaningless to compare it to 70/30 UPRO/TMF because it is not apples to apples. My portfolio is much more diversified than that and any differences in performance would be just as much due to differences in underlying allocation as it would be due to the method of implementation (which is the question currently at hand).
I agree, comparing it to that wouldn't be fair in that short time period.

Maybe it's more fair to compare your portfolio to a 3x leveraged ETF portfolio representing your asset allocation?
Maybe to something like UPRO/TMF/UGLD/etc...?
Perhaps. But either way, it is a short time period and not enough data to say whether my approach is superior/inferior. So that would then just take me back to pointing to the theoretical basis for my approach.
User avatar
Topic Author
tradri
Posts: 549
Joined: Tue Mar 23, 2021 5:42 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by tradri »

EfficientInvestor wrote: Mon Apr 12, 2021 3:51 pm Perhaps. But either way, it is a short time period and not enough data to say whether my approach is superior/inferior. So that would then just take me back to pointing to the theoretical basis for my approach.
Yeah, that's true. I wish you good luck. :beer
EfficientInvestor
Posts: 421
Joined: Thu Nov 01, 2018 7:02 pm
Location: Alabama

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by EfficientInvestor »

tradri wrote: Mon Apr 12, 2021 3:53 pm
EfficientInvestor wrote: Mon Apr 12, 2021 3:51 pm Perhaps. But either way, it is a short time period and not enough data to say whether my approach is superior/inferior. So that would then just take me back to pointing to the theoretical basis for my approach.
Yeah, that's true. I wish you good luck. :beer
Likewise! :beer
User avatar
Topic Author
tradri
Posts: 549
Joined: Tue Mar 23, 2021 5:42 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by tradri »

EfficientInvestor wrote: Mon Apr 12, 2021 9:30 am Why do you feel the need to reset the leverage so often? How often do your reset allocations on your existing portfolios? Why would you want to do the same thing that the Leveraged ETFs do under the hood? The daily reset of leverage is what leads to the volatility decay. I would say that the lack of rebalancing of a futures portfolio is a benefit to using futures. To help answer the question, I'll point to this backtest: https://www.portfoliovisualizer.com/bac ... on5_2=-200

It is 50/50 UPRO/TMF vs 150/150 SPY/TLT. The SPY/TLT portfolio represents the use of futures contracts to implement the strategy. It is rebalanced once per year. If I remove the LETFs from the backtest and go further back in time using VFINX/VUSTX instead of SPY/TLT, we get this: https://www.portfoliovisualizer.com/bac ... on5_2=-200

The point I'm trying to make with both backtests is that the portfolio has done just fine with only rebalancing once per year. As always, this backtest is no guarantee of the future. The futures-based portfolio is definitely prone to wild swings in overall exposure to the various assets in the portfolio, but isn't that the case with all investing? Just make sure your investment policy statement has rules for when to rebalance the same way you would a non-leveraged portfolio.
One thing I was wondering about, is whether one can simply model a futures portfolio by applying a perfect leverage factor in Portfolio Visualizer.

From what I have read, future contracts also have inherent costs in them, like the costs of holding it, as well as the rollover costs, right?
User avatar
cos
Posts: 401
Joined: Fri Aug 23, 2019 7:34 pm
Location: Boston
Contact:

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by cos »

tradri wrote: Mon Apr 12, 2021 5:03 pm
EfficientInvestor wrote: Mon Apr 12, 2021 9:30 am Why do you feel the need to reset the leverage so often? How often do your reset allocations on your existing portfolios? Why would you want to do the same thing that the Leveraged ETFs do under the hood? The daily reset of leverage is what leads to the volatility decay. I would say that the lack of rebalancing of a futures portfolio is a benefit to using futures. To help answer the question, I'll point to this backtest: https://www.portfoliovisualizer.com/bac ... on5_2=-200

It is 50/50 UPRO/TMF vs 150/150 SPY/TLT. The SPY/TLT portfolio represents the use of futures contracts to implement the strategy. It is rebalanced once per year. If I remove the LETFs from the backtest and go further back in time using VFINX/VUSTX instead of SPY/TLT, we get this: https://www.portfoliovisualizer.com/bac ... on5_2=-200

The point I'm trying to make with both backtests is that the portfolio has done just fine with only rebalancing once per year. As always, this backtest is no guarantee of the future. The futures-based portfolio is definitely prone to wild swings in overall exposure to the various assets in the portfolio, but isn't that the case with all investing? Just make sure your investment policy statement has rules for when to rebalance the same way you would a non-leveraged portfolio.
One thing I was wondering about, is whether one can simply model a futures portfolio by applying a perfect leverage factor in Portfolio Visualizer.

From what I have read, future contracts also have inherent costs in them, like the costs of holding it, as well as the rollover costs, right?
Yes, a negative allocation to CASHX does a good job approximating futures and their associated costs since it shorts the risk-free rate. See EfficientInvestor's links in the post you quoted for examples. The only cost missing from such backtests is taxation.
User avatar
Topic Author
tradri
Posts: 549
Joined: Tue Mar 23, 2021 5:42 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by tradri »

cos wrote: Mon Apr 12, 2021 5:28 pm
Yes, a negative allocation to CASHX does a good job approximating futures and their associated costs since it shorts the risk-free rate. See EfficientInvestor's links in the post you quoted for examples. The only cost missing from such backtests is taxation.
Right, it does subtract the risk free rate, but aren't there also rollover costs? Are these significant? Or are these just the costs of rebalancing?
User avatar
cos
Posts: 401
Joined: Fri Aug 23, 2019 7:34 pm
Location: Boston
Contact:

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by cos »

tradri wrote: Tue Apr 13, 2021 12:09 am
cos wrote: Mon Apr 12, 2021 5:28 pm
Yes, a negative allocation to CASHX does a good job approximating futures and their associated costs since it shorts the risk-free rate. See EfficientInvestor's links in the post you quoted for examples. The only cost missing from such backtests is taxation.
Right, it does subtract the risk free rate, but aren't there also rollover costs? Are these significant? Or are these just the costs of rebalancing?
Yes, no, yes. There are rollover costs, but if you're trading highly liquid futures like those on the S&P 500, these costs are insignificant. Indeed, these are just the costs of rebalancing (i.e. spreads).
User avatar
Topic Author
tradri
Posts: 549
Joined: Tue Mar 23, 2021 5:42 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by tradri »

cos wrote: Tue Apr 13, 2021 12:25 am
Yes, no, yes. There are rollover costs, but if you're trading highly liquid futures like those on the S&P 500, these costs are insignificant. Indeed, these are just the costs of rebalancing (i.e. spreads).
Ok, got it. So the ultimate question is what the best re-leveraging frequency is, right? Or more accurately put, how much leverage can you take with a quarterly/annually re-leveraged futures/options strategy, without facing the risk of total loss? And does this give you a higher CAGR than a pure 3x LETFs strategy?
dsasdg
Posts: 102
Joined: Sun Mar 21, 2021 4:24 am

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by dsasdg »

No, it is even better to buy some bitcoin.
User avatar
Topic Author
tradri
Posts: 549
Joined: Tue Mar 23, 2021 5:42 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by tradri »

dsasdg wrote: Tue Apr 13, 2021 2:47 am No, it is even better to buy some bitcoin.
LOL

Image
dsasdg
Posts: 102
Joined: Sun Mar 21, 2021 4:24 am

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by dsasdg »

tradri wrote: Tue Apr 13, 2021 3:24 am
dsasdg wrote: Tue Apr 13, 2021 2:47 am No, it is even better to buy some bitcoin.
LOL

Image
Long leverage etf will lower sharpe ratio/sortino ratio. It is better to buy some bitcoin because it can improve sortino ratio.
User avatar
Topic Author
tradri
Posts: 549
Joined: Tue Mar 23, 2021 5:42 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by tradri »

dsasdg wrote: Tue Apr 13, 2021 3:59 am
Long leverage etf will lower sharpe ratio/sortino ratio. It is better to buy some bitcoin because it can improve sortino ratio.
This isn't true if you leverage a portfolio of uncorrelated assets. Have a look at the backtest for 55/25/20 UPRO/TMF/UGLD and you will see that it has a higher Sharpe ratio than the S&P 500 with higher returns.
dsasdg
Posts: 102
Joined: Sun Mar 21, 2021 4:24 am

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by dsasdg »

tradri wrote: Tue Apr 13, 2021 4:03 am
dsasdg wrote: Tue Apr 13, 2021 3:59 am
Long leverage etf will lower sharpe ratio/sortino ratio. It is better to buy some bitcoin because it can improve sortino ratio.
This isn't true if you leverage a portfolio of uncorrelated assets. Have a look at the backtest for 55/25/20 UPRO/TMF/UGLD and you will see that it has a higher Sharpe ratio than the S&P 500 with higher returns.
It is better to hold SPY/TLT/GLD
Last edited by dsasdg on Tue Apr 13, 2021 5:05 am, edited 1 time in total.
User avatar
Topic Author
tradri
Posts: 549
Joined: Tue Mar 23, 2021 5:42 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by tradri »

dsasdg wrote: Tue Apr 13, 2021 4:56 am
It is better to hold spy/tlt/glad.
Yes, if you are only looking for the highest Sharpe ratio, leveraging isn't going to help. I am interested in the highest CAGR though.
User avatar
watchnerd
Posts: 8327
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by watchnerd »

Once the 10 YR hits 2%, maybe I'll pop $50k into 70/30 UPRO/TMF.
60% Global Market Stocks (VT,FM) | 38% Global Market Bonds | 2% Alts || LMP TIPS/STRIPS || RSU + ESPP
User avatar
Topic Author
tradri
Posts: 549
Joined: Tue Mar 23, 2021 5:42 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by tradri »

watchnerd wrote: Tue Apr 13, 2021 7:48 pm Once the 10 YR hits 2%, maybe I'll pop $50k into 70/30 UPRO/TMF.
Ok. I think the cool thing about holding good, uncorrelated assets is that they should deliver satisfactory results in any market environment.
User avatar
watchnerd
Posts: 8327
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by watchnerd »

tradri wrote: Wed Apr 14, 2021 4:54 am
watchnerd wrote: Tue Apr 13, 2021 7:48 pm Once the 10 YR hits 2%, maybe I'll pop $50k into 70/30 UPRO/TMF.
Ok. I think the cool thing about holding good, uncorrelated assets is that they should deliver satisfactory results in any market environment.
Well, you're ignoring that we're in a rising interest rate environment and what that will do to long Treasuries, especially geared ones.

in any case, I changed my mind.

Instead, as of today, I'm going 50% VT / 29% UPRO / 15% EURL / 6% EDC (which is basically a geared version of my main stock allocation in my sig) with the measly $70k in my Roth.

I'll rebalance quarterly.

If nothing else, it should provide a good funding pool for FIRE medical expenses and something to pass on to heirs.
60% Global Market Stocks (VT,FM) | 38% Global Market Bonds | 2% Alts || LMP TIPS/STRIPS || RSU + ESPP
User avatar
Topic Author
tradri
Posts: 549
Joined: Tue Mar 23, 2021 5:42 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by tradri »

watchnerd wrote: Wed Apr 14, 2021 8:19 am
Well, you're ignoring that we're in a rising interest rate environment and what that will do to long Treasuries, especially geared ones.

in any case, I changed my mind.

Instead, as of today, I'm going 50% VT / 29% UPRO / 15% EURL / 6% EDC (which is basically a geared version of my main stock allocation in my sig) with the measly $70k in my Roth.

I'll rebalance quarterly.

If nothing else, it should provide a good funding pool for FIRE medical expenses and something to pass on to heirs.
You probably already know this, but adding uncorrelated assets to a leveraged ETF strategy actually improves the CAGR according to the backtests. Your portfolio only focuses on one asset class, so the correlation between those ETFs is going to be high.

I agree that rising interest rates/inflation is a risk, which is why I plan to add 3x gold to save the portfolio in such a market environment, like the one from the 1970s to the 1980s. (Unfortunately such an ETF isn't available in the US though)

Anyways, I wish you good luck. :beer
User avatar
watchnerd
Posts: 8327
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by watchnerd »

tradri wrote: Wed Apr 14, 2021 8:35 am Your portfolio only focuses on one asset class, so the correlation between those ETFs is going to be high.
No, it doesn't.

Look at my sig.
60% Global Market Stocks (VT,FM) | 38% Global Market Bonds | 2% Alts || LMP TIPS/STRIPS || RSU + ESPP
User avatar
Topic Author
tradri
Posts: 549
Joined: Tue Mar 23, 2021 5:42 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by tradri »

watchnerd wrote: Wed Apr 14, 2021 9:22 am No, it doesn't.

Look at my sig.
Oh, ok. I was talking about the portfolio you stated above.
watchnerd wrote: Wed Apr 14, 2021 8:19 am Instead, as of today, I'm going 50% VT / 29% UPRO / 15% EURL / 6% EDC (which is basically a geared version of my main stock allocation in my sig) with the measly $70k in my Roth.
Hydromod
Posts: 535
Joined: Tue Mar 26, 2019 10:21 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by Hydromod »

watchnerd wrote: Wed Apr 14, 2021 9:22 am
tradri wrote: Wed Apr 14, 2021 8:35 am Your portfolio only focuses on one asset class, so the correlation between those ETFs is going to be high.
No, it doesn't.

Look at my sig.
One of the limitations of a Roth is that you can't transfer into it in order to rebalance, and transferring out is a one-way street. So in effect you really are accepting the risk of 3x equities in isolation from the rest of the portfolio with your strategy.

Just something to consider.
User avatar
watchnerd
Posts: 8327
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by watchnerd »

Hydromod wrote: Wed Apr 14, 2021 9:37 am
watchnerd wrote: Wed Apr 14, 2021 9:22 am
tradri wrote: Wed Apr 14, 2021 8:35 am Your portfolio only focuses on one asset class, so the correlation between those ETFs is going to be high.
No, it doesn't.

Look at my sig.
One of the limitations of a Roth is that you can't transfer into it in order to rebalance, and transferring out is a one-way street. So in effect you really are accepting the risk of 3x equities in isolation from the rest of the portfolio with your strategy.

Just something to consider.
Well, the Roth route is now off the table.

It's at Vanguard and Vanguard apparently won't let you buy leveraged ETFs in any account.

So it will have to be another account elsewhere.
60% Global Market Stocks (VT,FM) | 38% Global Market Bonds | 2% Alts || LMP TIPS/STRIPS || RSU + ESPP
User avatar
Topic Author
tradri
Posts: 549
Joined: Tue Mar 23, 2021 5:42 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by tradri »

Hydromod wrote: Wed Apr 14, 2021 9:37 am
One of the limitations of a Roth is that you can't transfer into it in order to rebalance, and transferring out is a one-way street. So in effect you really are accepting the risk of 3x equities in isolation from the rest of the portfolio with your strategy.

Just something to consider.
Excuse my non-US noob question, but how do you fund a Roth account if you can't transfer into it?
Hydromod
Posts: 535
Joined: Tue Mar 26, 2019 10:21 pm

Re: Are 3x leveraged ETFs the long-term winning strategy?

Post by Hydromod »

tradri wrote: Wed Apr 14, 2021 9:47 am
Hydromod wrote: Wed Apr 14, 2021 9:37 am
One of the limitations of a Roth is that you can't transfer into it in order to rebalance, and transferring out is a one-way street. So in effect you really are accepting the risk of 3x equities in isolation from the rest of the portfolio with your strategy.

Just something to consider.
Excuse my non-US noob question, but how do you fund a Roth account if you can't transfer into it?
You are allowed to have a Roth IRA that you can (i) directly fund up to $6000/yr ($7000/yr over 50), if you earn less than a certain threshold, or (ii) indirectly fund up to the same amount by creating a traditional IRA and rolling it over (the backdoor Roth), if you make more than the threshold. The backdoor process has certain limitations if you have other traditional IRAs.

You can also have a 401(k) or 403(b) Roth, which is contributed to like a normal 401(k) except with after-tax income instead of before-tax income.
Post Reply