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

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
DMoogle
Posts: 152
Joined: Sat Oct 31, 2020 10:24 pm

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

Post by DMoogle »

RovenSkyfall wrote: Wed Apr 07, 2021 10:08 amWhat was the return of the alternative?
That completely misses the point.

Bit Digital (BTBT) was the best performing stock in 2020 according to a quick Google search, returning 3,691%. That's an alternative. May as well put all your money in that, right? Or at least a portfolio of the top 10 returning stocks to mitigate risk right?

Of course not, because past results are not necessarily indicative of expected return. Expected return = what this investment should deliver in the long-run. This entire thread has been a debate on what that expected return is for LETFs.
User avatar
Topic Author
tradri
Posts: 386
Joined: Tue Mar 23, 2021 5:42 pm

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

Post by tradri »

RovenSkyfall wrote: Wed Apr 07, 2021 10:19 am
I have read through the thread before. What are you referencing? If you can clearly express your concern people might be able to respond better.
I'm referencing the simulations they have done in that thread that conclude that UPRO didn't really deliver higher CAGRs than the S&P 500.
User avatar
RovenSkyfall
Posts: 216
Joined: Wed Apr 01, 2020 11:40 am

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

Post by RovenSkyfall »

DMoogle wrote: Wed Apr 07, 2021 10:25 am
RovenSkyfall wrote: Wed Apr 07, 2021 10:08 amWhat was the return of the alternative?
That completely misses the point.

Bit Digital (BTBT) was the best performing stock in 2020 according to a quick Google search, returning 3,691%. That's an alternative. May as well put all your money in that, right? Or at least a portfolio of the top 10 returning stocks to mitigate risk right?

Of course not, because past results are not indicative of expected return. Expected return = what this investment should deliver in the long-run. This entire thread has been a debate on what that expected return is for LETFs.
It does not miss the point. When an investor considers an investment, it is always in comparison to an alternative (even if that is cash). That is the choice investors must make. This choice includes variable such as the expected return as well as other considerations (risk).

14.24% is much better than cash, but I am guessing that is not what you are comparing it to. Usually people would compare the LETF to the non-leveraged benchmark like the SP500. Then you would compare the return (both went through a bull) and risks. If there is excess return, then it would have a higher expected return, albeit with a higher risk. Usually, when speaking about expected return, the average return is used (not the geometric). For the available real data, UPRO had an artimetic mean of 45% vs SPY's 15%, so the expected return is higher.

I have no attachment to the outcome of the discussion, but hope it arrives at a conclusion based on sound reasoning. The probability of outcomes for a 3xLETF of the SP500 is not great over a 20 year period, mostly due to the tail risk over a long period of time. Likely for this reason, I have not seen any convincing argument for a long term B&H of UPRO alone. That does not mean that you can't have higher expected returns with a combination of LETFs compared to a TSM fund.
I saved my money, but it can't save me | The Chariot
User avatar
RovenSkyfall
Posts: 216
Joined: Wed Apr 01, 2020 11:40 am

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

Post by RovenSkyfall »

tradri wrote: Wed Apr 07, 2021 11:03 am
RovenSkyfall wrote: Wed Apr 07, 2021 10:19 am
I have read through the thread before. What are you referencing? If you can clearly express your concern people might be able to respond better.
I'm referencing the simulations they have done in that thread that conclude that UPRO didn't really deliver higher CAGRs than the S&P 500.
I don't remember that conclusion being discussed. Can you link the post you are referencing?
I saved my money, but it can't save me | The Chariot
User avatar
Topic Author
tradri
Posts: 386
Joined: Tue Mar 23, 2021 5:42 pm

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

Post by tradri »

RovenSkyfall wrote: Wed Apr 07, 2021 12:05 pm I don't remember that conclusion being discussed. Can you link the post you are referencing?
Here are the latest results: viewtopic.php?p=5822849#p5822849

UPRO produced 11.54% CAGR from 1955 to 2020.
40/60 UPRO/TMF produced 12.43% CAGR from 1955 to 2020. (without rebalancing costs, I assume)
User avatar
RovenSkyfall
Posts: 216
Joined: Wed Apr 01, 2020 11:40 am

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

Post by RovenSkyfall »

tradri wrote: Wed Apr 07, 2021 12:23 pm
RovenSkyfall wrote: Wed Apr 07, 2021 12:05 pm I don't remember that conclusion being discussed. Can you link the post you are referencing?
Here are the latest results: viewtopic.php?p=5822849#p5822849

UPRO produced 11.54% CAGR from 1955 to 2020.
40/60 UPRO/TMF produced 12.43% CAGR from 1955 to 2020. (without rebalancing costs, I assume)
I will have to look when I get home as it is taking too long to load now. Two quick points are that currently the HFEA is generally run at 55/45 UPRO/TMF and it is quarterly rebalanced (if you want to compare what most people are using).
I saved my money, but it can't save me | The Chariot
User avatar
Topic Author
tradri
Posts: 386
Joined: Tue Mar 23, 2021 5:42 pm

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

Post by tradri »

RovenSkyfall wrote: Wed Apr 07, 2021 12:47 pm
I will have to look when I get home as it is taking too long to load now. Two quick points are that currently the HFEA is generally run at 55/45 UPRO/TMF and it is quarterly rebalanced (if you want to compare what most people are using).
55/45 UPRO/TMF produces a better CAGR, at 13.77%.

Still, I assume the quarterly rebalancing costs aren't negligible.

According to these backtests, the best UPRO/TMF combination is 70/30 UPRO/TMF, producing a CAGR of 14.31%.

P.S. After playing around with that spreadsheet, this Hedgefundie risk parity strategy really does sound enticing. One question though: Why exactly 55/45 UPRO/TMF and not 70/30 UPRO/TMF, when the latter produces a higher CAGR?
Rogue_trader
Posts: 19
Joined: Wed Apr 07, 2021 1:14 pm

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

Post by Rogue_trader »

Dear all,

I fully understand that the returns of leveraged ETFs have to reflect at least the following costs:
1) management fee - included in the expense ratio
2) borrowing costs - floating leg of the total return equity swaps pays (probably) 1M LIBOR + spread
3) other costs - this includes transaction fees, slippage, bid-ask spreads.. They are probably very low.

Neglecting 3) I end up with the following equation describing the relationship between the returns of the underlying asset (or index) and the leveraged ETF:

LETF = L*UNDERLYING - (L-1)*1M_LIBOR - SWAP*Wswap - EXPESNE_RATIO

Yet, this is not at all what I observe in case of UPRO, SSO, TQQQ.. Somehow all these ETFs beat the underlying index multiplied by the leverage. So in other words:
LETF => L*UNDERLYING
and all the costs are covered. So UPRO, SSO, TQQQ beat the simple index multiplied by the leverage ratio.

How is it possible? :confused Any ideas? :confused

Here you can see a chart of UPRO vs 3*S&P500.
https://postimg.cc/w7DK5sCx

Here you can see a telltale chart of UPRO vs 3*S&P500.
https://postimg.cc/jWndbjWD

Do you have any idea how it is possible that UPRO, SSO, TQQQ... beat 3*S&P500 (and I mean here total return S&P500 with reinvested dividends).

Thank you very much!
User avatar
OohLaLa
Posts: 141
Joined: Tue Mar 09, 2021 8:26 pm

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

Post by OohLaLa »

tradri wrote: Wed Apr 07, 2021 1:04 pm [...]
Still, I assume the quarterly rebalancing costs aren't negligible.
I would say that they are for many people, considering how low fees are becoming. Just to give you an idea, for me, rebalancing a two-fund portfolio would cost me 5-10 (max) bucks per quarter = 20-40 bucks per year. This absolutely is negligible, with any allocation that isn't symbolic in amount.
tradri wrote: Wed Apr 07, 2021 1:04 pm According to these backtests, the best UPRO/TMF combination is 70/30 UPRO/TMF, producing a CAGR of 14.31%.

P.S. After playing around with that spreadsheet, this Hedgefundie risk parity strategy really does sound enticing. One question though: Why exactly 55/45 UPRO/TMF and not 70/30 UPRO/TMF, when the latter produces a higher CAGR?
I would assume it would be because 70% 3x equity is too high, in terms of potential drawdowns. I am currently at that level, but I definitely am decreasing that yearly.
User avatar
Topic Author
tradri
Posts: 386
Joined: Tue Mar 23, 2021 5:42 pm

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

Post by tradri »

OohLaLa wrote: Wed Apr 07, 2021 1:49 pm
tradri wrote: Wed Apr 07, 2021 1:04 pm [...]
Still, I assume the quarterly rebalancing costs aren't negligible.
I would say that they are for many people, considering how low fees are becoming. Just to give you an idea, for me, rebalancing a two-fund portfolio would cost me 5-10 (max) bucks per quarter = 20-40 bucks per year. This absolutely is negligible, with any allocation that isn't symbolic in amount.
tradri wrote: Wed Apr 07, 2021 1:04 pm According to these backtests, the best UPRO/TMF combination is 70/30 UPRO/TMF, producing a CAGR of 14.31%.

P.S. After playing around with that spreadsheet, this Hedgefundie risk parity strategy really does sound enticing. One question though: Why exactly 55/45 UPRO/TMF and not 70/30 UPRO/TMF, when the latter produces a higher CAGR?
I would assume it would be because 70% 3x equity is too high, in terms of potential drawdowns. I am currently at that level, but I definitely am decreasing that yearly.
If the average long-term CAGR is 14.31%, then occasionally rebalancing 2 funds in a portfolio is definitely worth it. (The CAGR is most likely even higher than that, when the funds are rebalanced quarterly, as the Hedgefundie strategy suggests)

Assuming one is only concerned about maximizing the CAGR (which is the whole premise of this thread), then a 70/30 UPRO/TMF allocation would be close to ideal, right?
chris319
Posts: 262
Joined: Thu Jan 28, 2021 6:04 pm

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

Post by chris319 »

If the average long-term CAGR is 14.31%, then occasionally rebalancing 2 funds in a portfolio is definitely worth it.
Don't forget to consider the tax consequences of rebalancing.
GOMEZ: Morticia! Consolidated Fuzz just hit 212 — I sold it for 6 — saved all that tax. MORTICIA: Brilliant!
User avatar
Topic Author
tradri
Posts: 386
Joined: Tue Mar 23, 2021 5:42 pm

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

Post by tradri »

chris319 wrote: Wed Apr 07, 2021 2:25 pm
If the average long-term CAGR is 14.31%, then occasionally rebalancing 2 funds in a portfolio is definitely worth it.
Don't forget to consider the tax consequences of rebalancing.
True, but the average boglehead also holds a portfolio of multiple funds, so everybody has that same burden.

In the accumulation phase, the rebalancing can be reduced by just buying the fund that is under weighted.
DMoogle
Posts: 152
Joined: Sat Oct 31, 2020 10:24 pm

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

Post by DMoogle »

chris319 wrote: Wed Apr 07, 2021 2:25 pm
If the average long-term CAGR is 14.31%, then occasionally rebalancing 2 funds in a portfolio is definitely worth it.
Don't forget to consider the tax consequences of rebalancing.
Yes, very important. I assume that might be what tradri was referring to in terms of rebalancing costs, but that's exactly why the HFEA thread consistently stresses that it's only a strategy for tax-advantaged accounts.

@RovenSkyfall: I owe you a response, but want to read through what you linked first.
User avatar
Topic Author
tradri
Posts: 386
Joined: Tue Mar 23, 2021 5:42 pm

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

Post by tradri »

DMoogle wrote: Wed Apr 07, 2021 2:31 pm Yes, very important. I assume that might be what tradri was referring to in terms of rebalancing costs, but that's exactly why the HFEA thread consistently stresses that it's only a strategy for tax-advantaged accounts.

@RovenSkyfall: I owe you a response, but want to read through what you linked first.
Can you explain to me, why the rebalancing costs should be a bigger problem for this strategy, than for the average Three Fund Portfolio?
DMoogle
Posts: 152
Joined: Sat Oct 31, 2020 10:24 pm

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

Post by DMoogle »

tradri wrote: Wed Apr 07, 2021 2:33 pmCan you explain to me, why the rebalancing costs should be a bigger problem for this strategy, than for the average Three Fund Portfolio?
Hmmm, not 100% sure. I've always kind of done my own thing and have avoided rebalancing entirely in my taxable accounts (except for via contributions).

One thing that would probably make a substantial difference: the fund ratios between UPRO and TMF are MUCH more likely to substantially deviate from their original allocation just due to the volatility than the three-fund portfolio will, so more rebalancing will be required.
User avatar
OohLaLa
Posts: 141
Joined: Tue Mar 09, 2021 8:26 pm

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

Post by OohLaLa »

tradri wrote: Wed Apr 07, 2021 1:57 pm Assuming one is only concerned about maximizing the CAGR (which is the whole premise of this thread), then a 70/30 UPRO/TMF allocation would be close to ideal, right?
If someone just wants max CAGR, at any cost, it's as simple as backtesting with small changes in AA until they find it. From what I recall when checking, some addition of TMF will definitely allow to outperform UPRO alone, even when only looking at CAGR. The volatility for any 3x fund alone is that bad.

Is anybody actually still suggesting going all-in on a single 3x fund? I get 1.25x or maybe 1.5x if you want to push your luck but 2x-3x?!?! You really are betting that the 5-10 last years (minus 2020) will keep repeating over and over again. The moment you hit any headwind, a blend of UPRO + TMF or equivalent wins hands down.
User avatar
Topic Author
tradri
Posts: 386
Joined: Tue Mar 23, 2021 5:42 pm

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

Post by tradri »

DMoogle wrote: Wed Apr 07, 2021 2:46 pm Hmmm, not 100% sure. I've always kind of done my own thing and have avoided rebalancing entirely in my taxable accounts (except for via contributions).

One thing that would probably make a substantial difference: the fund ratios between UPRO and TMF are MUCH more likely to substantially deviate from their original allocation just due to the volatility than the three-fund portfolio will, so more rebalancing will be required.
Yes, it will deviate more, but if I'm not mistaken the backtest spreadsheet only rebalances once per year, and even then it produces a significantly higher CAGR.

I tested the 70/30 UPRO/TMF combination since inception in Portfolio Visualizer, and the CAGR significantly improved when rebalanced quarterly.
User avatar
OohLaLa
Posts: 141
Joined: Tue Mar 09, 2021 8:26 pm

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

Post by OohLaLa »

DMoogle wrote: Wed Apr 07, 2021 2:46 pm
tradri wrote: Wed Apr 07, 2021 2:33 pmCan you explain to me, why the rebalancing costs should be a bigger problem for this strategy, than for the average Three Fund Portfolio?
Hmmm, not 100% sure. I've always kind of done my own thing and have avoided rebalancing entirely in my taxable accounts (except for via contributions).

One thing that would probably make a substantial difference: the fund ratios between UPRO and TMF are MUCH more likely to substantially deviate from their original allocation just due to the volatility than the three-fund portfolio will, so more rebalancing will be required.
Yeah, I don't get all the worry about taxable accounts and rebalancing. If you are really comitting to some strategy/ AA, then:
===During accumulation years: rebalance with new funds.
===During withdrawal years: use your quarterly withdrawals, to finance your lifestyle as well as a way to rebalance.

This can minimize "waste" in good part. And, like you said, it's worse with the 3x funds, due to volatility. Hopefully, people don't stick with 3x strategies in their withdrawal phase.
User avatar
Topic Author
tradri
Posts: 386
Joined: Tue Mar 23, 2021 5:42 pm

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

Post by tradri »

OohLaLa wrote: Wed Apr 07, 2021 2:50 pm If someone just wants max CAGR, at any cost, it's as simple as backtesting with small changes in AA until they find it. From what I recall when checking, some addition of TMF will definitely allow to outperform UPRO alone, even when only looking at CAGR. The volatility for any 3x fund alone is that bad.

Is anybody actually still suggesting going all-in on a single 3x fund? I get 1.25x or maybe 1.5x if you want to push your luck but 2x-3x?!?! You really are betting that the 5-10 last years (minus 2020) will keep repeating over and over again. The moment you hit any headwind, a blend of UPRO + TMF or equivalent wins hands down.
After reading through the whole Simulating Returns of Leveraged ETFs thread and seeing for myself, that higher overnight LIBOR rates have a significant negative effect on UPRO outperforming long-term, I acknowledge than 100% UPRO isn't the wisest investment to maximize the CAGR long-term.

I played around with the backtest spreadsheet, and from what I could tell the highest CAGR is achieved with 70/30 UPRO/TMF.
User avatar
Topic Author
tradri
Posts: 386
Joined: Tue Mar 23, 2021 5:42 pm

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

Post by tradri »

OohLaLa wrote: Wed Apr 07, 2021 2:54 pm Yeah, I don't get all the worry about taxable accounts and rebalancing. If you are really comitting to some strategy/ AA, then:
===During accumulation years: rebalance with new funds.
===During withdrawal years: use your quarterly withdrawals, to finance your lifestyle as well as a way to rebalance.

This can minimize "waste" in good part. And, like you said, it's worse with the 3x funds, due to volatility. Hopefully, people don't stick with 3x strategies in their withdrawal phase.
I agree that owning a portfolio that is more volatile than the S&P 500 is probably not a good idea when you are trying to live off that money.
User avatar
Ramjet
Posts: 802
Joined: Thu Feb 06, 2020 11:45 am
Location: Ohio

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

Post by Ramjet »

tradri wrote: Wed Apr 07, 2021 1:04 pm P.S. After playing around with that spreadsheet, this Hedgefundie risk parity strategy really does sound enticing. One question though: Why exactly 55/45 UPRO/TMF and not 70/30 UPRO/TMF, when the latter produces a higher CAGR?
Drawdown difference between allocations not worth the extra CAGR
"Every normal man must be tempted, at times, to spit upon his hands, hoist the black flag, and begin slitting throats" H.L. Mencken
User avatar
Topic Author
tradri
Posts: 386
Joined: Tue Mar 23, 2021 5:42 pm

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

Post by tradri »

Ramjet wrote: Wed Apr 07, 2021 3:22 pm
Drawdown difference between allocations not worth the extra CAGR
Why not? What's the problem with taking on a higher volatility?
chris319
Posts: 262
Joined: Thu Jan 28, 2021 6:04 pm

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

Post by chris319 »

50% UPRO and 50% SSO actually performs better than 100% UPRO. This gives you 2.5x leverage. It is easy to test this.
GOMEZ: Morticia! Consolidated Fuzz just hit 212 — I sold it for 6 — saved all that tax. MORTICIA: Brilliant!
alex_686
Posts: 8131
Joined: Mon Feb 09, 2015 2:39 pm

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

Post by alex_686 »

So, I am late to the game and I have not read the whole thread, but let us see if I can help.
tradri wrote: Wed Apr 07, 2021 3:19 pm I don't know if this is the reason why your simulations underperform, but from what I can tell one should use the overnight rate, not the 1 month LIBOR.
and
Rogue_trader wrote: Wed Apr 07, 2021 4:45 pm
tradri wrote: Wed Apr 07, 2021 4:17 pm
Rogue_trader wrote: Wed Apr 07, 2021 3:14 pm LETF = L*UNDERLYING or actually LETF > L*UNDERLYING
Did you just multiply the S&P 500 by 3x over that whole time period?

Because due to the daily compounding, a 3x leveraged ETF produces more than 3x leverage in a strong bull market, such as the one we have seen in the past few years.

Anyways, here is the Python code I used to estimate CAGRs: https://pastebin.com/qxqQV5xg
I am aware of the volatility drag.. I multiplied daily returns of S&P500 by 3 just like UPRO aims to do! UPRO should be lagging behind 3*S&P500 and not beating it. It should be lagging by expense ratio, borrowing costs and possibly also other costs (transactions fees, slippage...), but it is not the case. I really do not understand why at all. It makes no sense, because it means that there is an arbitrage opportunity.
So a couple of points.

The generalized formula for pricing for future is: Futures Price = Spot Price (1 + Risk Free Rate - Dividend Rate)

I forget what the average duration until maturity of the futures that UPRO use, but I think they are in the 6 to 9 month range. It, of course, varies daily.

Next, have you factored in dividends into your calculations? Somebody investing directly in the S&P would get them, somebody investing in futures would not. Futures are calculated off of the price index, not the total return index.

Next, it is not volatility drag. It is something subtler, but I forget the exact mathematical term. The issue is that UPRO rebalances daily. As such returns are highly path dependent. There are fact patters where the index is up but UPRO would be down. IIRC, historically after about 6 weeks the explanatory power of S&P 500 *3 for UPRO is about 50%. And declines after than. If you are going to use UPRO as a core holding you are going to need to rebalance & rehedge monthly.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
alex_686
Posts: 8131
Joined: Mon Feb 09, 2015 2:39 pm

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

Post by alex_686 »

tradri wrote: Wed Apr 07, 2021 3:24 pm
Ramjet wrote: Wed Apr 07, 2021 3:22 pm
Drawdown difference between allocations not worth the extra CAGR
Why not? What's the problem with taking on a higher volatility?
IIRC, CAGR = Return - standard deviation of returns.

As you crank up leverage you increase both returns and the standard deviation. At first CAGR increases due to the positive leverage of returns. however at some point the volatility drag will become greater and CAGR will fall.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
chris319
Posts: 262
Joined: Thu Jan 28, 2021 6:04 pm

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

Post by chris319 »

GOMEZ: Morticia! Consolidated Fuzz just hit 212 — I sold it for 6 — saved all that tax. MORTICIA: Brilliant!
chris319
Posts: 262
Joined: Thu Jan 28, 2021 6:04 pm

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

Post by chris319 »

Ralph Vince explains in his books why too much leverage is counterproductive, i.e. negatively impacts CAGR. He goes into a great deal of mathematical theory.

An oversimplified explanation is that, given a sequence of periodic returns, the drawdowns can be so deep that the positive returns cannot fully compensate for them, resulting in diminished CAGR. I have thus found that a blend of 50/50 2x and 3x to be a sweet spot.

This has its roots in the Kelly criterion.
GOMEZ: Morticia! Consolidated Fuzz just hit 212 — I sold it for 6 — saved all that tax. MORTICIA: Brilliant!
Rogue_trader
Posts: 19
Joined: Wed Apr 07, 2021 1:14 pm

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

Post by Rogue_trader »

alex_686 wrote: Wed Apr 07, 2021 4:57 pm So, I am late to the game and I have not read the whole thread, but let us see if I can help.
tradri wrote: Wed Apr 07, 2021 3:19 pm I don't know if this is the reason why your simulations underperform, but from what I can tell one should use the overnight rate, not the 1 month LIBOR.
and
Rogue_trader wrote: Wed Apr 07, 2021 4:45 pm
tradri wrote: Wed Apr 07, 2021 4:17 pm
Rogue_trader wrote: Wed Apr 07, 2021 3:14 pm LETF = L*UNDERLYING or actually LETF > L*UNDERLYING
Did you just multiply the S&P 500 by 3x over that whole time period?

Because due to the daily compounding, a 3x leveraged ETF produces more than 3x leverage in a strong bull market, such as the one we have seen in the past few years.

Anyways, here is the Python code I used to estimate CAGRs: https://pastebin.com/qxqQV5xg
I am aware of the volatility drag.. I multiplied daily returns of S&P500 by 3 just like UPRO aims to do! UPRO should be lagging behind 3*S&P500 and not beating it. It should be lagging by expense ratio, borrowing costs and possibly also other costs (transactions fees, slippage...), but it is not the case. I really do not understand why at all. It makes no sense, because it means that there is an arbitrage opportunity.
So a couple of points.

The generalized formula for pricing for future is: Futures Price = Spot Price (1 + Risk Free Rate - Dividend Rate)

I forget what the average duration until maturity of the futures that UPRO use, but I think they are in the 6 to 9 month range. It, of course, varies daily.

Next, have you factored in dividends into your calculations? Somebody investing directly in the S&P would get them, somebody investing in futures would not. Futures are calculated off of the price index, not the total return index.

Next, it is not volatility drag. It is something subtler, but I forget the exact mathematical term. The issue is that UPRO rebalances daily. As such returns are highly path dependent. There are fact patters where the index is up but UPRO would be down. IIRC, historically after about 6 weeks the explanatory power of S&P 500 *3 for UPRO is about 50%. And declines after than. If you are going to use UPRO as a core holding you are going to need to rebalance & rehedge monthly.
Yes, the formula is correct and as you can see there is risk free rate. The same risk free rate should be also emended in UPRO, because UPRO cannot borrow for free, but it seems that it can and not only that they are even borrowing at negative rates. Or there must be some other source of return boost, which makes UPRO to beat 3*S&P500.

Dividends are irrelevant in this case. I am using S&P500 total return index and UPRO is most likely also based on that (if not then it is even less logical why it beats 3*S&P500).

Volatility drag is totally irrelevant here. I am talking about daily returns and their cumulative product not annual returns! Have you tried to compare UPRO and 3*S&P500? You can easily do it in excel. UPRO beats it and it beats it big.
MotoTrojan
Posts: 11017
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan »

tradri wrote: Wed Apr 07, 2021 6:11 am After reading through all the arguments for and against leverage, it seems to me that all the hassle isn't worth it.

Convincing backtests have shown, that using leveraged ETFs like UPRO doesn't produce superior results long-term.

Guess I will have to stick to 100% stocks to get the maximum reliable return long-term. :sharebeer
Proud of you. Now come to the dark-side and get into deep-factor tilts!
MotoTrojan
Posts: 11017
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan »

Rogue_trader wrote: Wed Apr 07, 2021 5:54 pm

Volatility drag is totally irrelevant here. I am talking about daily returns and their cumulative product not annual returns! Have you tried to compare UPRO and 3*S&P500? You can easily do it in excel. UPRO beats it and it beats it big.
I think you need to bust out excel a bit more as it apparently isn't as easy as you think. Are you doing 3x returns or 3x the final value? That isn't how compounding works...

3x S&P500 isn't the final wealth gained of S&P500 multiplied by 3x, it is 3x the CAGR, which with linear returns over many years would be WAY more than 3x the wealth, in fact taken to infinity it would be an infinite times more wealth gained.

See here, even with the negative drag of -200% CASHX (so a bit less than 3x S&P500) it is actually UPRO that GETS beat (3x S&P500 has a 43% CAGR while UPRO got a 36.4% CAGR).

https://www.portfoliovisualizer.com/bac ... on3_2=-200

Other than the expense ratio, volatility drag is the ONLY relevant reason why UPRO underperformed 3x S&P500. This is an incredibly flawed take.
chris319
Posts: 262
Joined: Thu Jan 28, 2021 6:04 pm

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

Post by chris319 »

The same risk free rate should be also emended in UPRO, because UPRO cannot borrow for free, but it seems that it can and not only that they are even borrowing at negative rates. Or there must be some other source of return boost, which makes UPRO to beat 3*S&P500.
What exactly is the fund borrowing? What kind of asset?

We are well aware that these funds promise 2x or 3x the daily return, not annual. Some people don't understand this but we do!
GOMEZ: Morticia! Consolidated Fuzz just hit 212 — I sold it for 6 — saved all that tax. MORTICIA: Brilliant!
MotoTrojan
Posts: 11017
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan »

chris319 wrote: Wed Apr 07, 2021 6:12 pm
The same risk free rate should be also emended in UPRO, because UPRO cannot borrow for free, but it seems that it can and not only that they are even borrowing at negative rates. Or there must be some other source of return boost, which makes UPRO to beat 3*S&P500.
What exactly is the fund borrowing? What kind of asset?

We are well aware that these funds promise 2x or 3x the daily return, not annual. Some people don't understand this but we do!
So you think it is free magic leverage? Come on...

It is called a swap.

https://www.investopedia.com/terms/s/sw ... in%20swaps.
Total Return Swaps
In a total return swap, the total return from an asset is exchanged for a fixed interest rate. This gives the party paying the fixed-rate exposure to the underlying asset—a stock or an index. For example, an investor could pay a fixed rate to one party in return for the capital appreciation plus dividend payments of a pool of stocks.
You say "we" understand, but the post you quoted is saying an outright lie that UPRO outperformed 3x S&P500.
chris319
Posts: 262
Joined: Thu Jan 28, 2021 6:04 pm

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

Post by chris319 »

So you think it is free magic leverage? Come on...
No, I do not think it is "free magic leverage" and never said I did. I corrected the misconception in the post, if you picked up on it.

I wasn't familiar with how swaps work. Thanks for illuminating.
GOMEZ: Morticia! Consolidated Fuzz just hit 212 — I sold it for 6 — saved all that tax. MORTICIA: Brilliant!
alex_686
Posts: 8131
Joined: Mon Feb 09, 2015 2:39 pm

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

Post by alex_686 »

Rogue_trader wrote: Wed Apr 07, 2021 5:54 pm Dividends are irrelevant in this case. I am using S&P500 total return index and UPRO is most likely also based on that (if not then it is even less logical why it beats 3*S&P500).
Having done this for a living, having looked at the prospectus a few years ago, UPRO trades S&P Futures contracts on the CME. These use the S&P price index - so no dividends.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
alex_686
Posts: 8131
Joined: Mon Feb 09, 2015 2:39 pm

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

Post by alex_686 »

MotoTrojan wrote: Wed Apr 07, 2021 6:00 pm
Rogue_trader wrote: Wed Apr 07, 2021 5:54 pm

Volatility drag is totally irrelevant here. I am talking about daily returns and their cumulative product not annual returns! Have you tried to compare UPRO and 3*S&P500? You can easily do it in excel. UPRO beats it and it beats it big.
I think you need to bust out excel a bit more as it apparently isn't as easy as you think. Are you doing 3x returns or 3x the final value? That isn't how compounding works...

3x S&P500 isn't the final wealth gained of S&P500 multiplied by 3x, it is 3x the CAGR, which with linear returns over many years would be WAY more than 3x the wealth, in fact taken to infinity it would be an infinite times more wealth gained.

See here, even with the negative drag of -200% CASHX (so a bit less than 3x S&P500) it is actually UPRO that GETS beat (3x S&P500 has a 43% CAGR while UPRO got a 36.4% CAGR).

https://www.portfoliovisualizer.com/bac ... on3_2=-200

Other than the expense ratio, volatility drag is the ONLY relevant reason why UPRO underperformed 3x S&P500. This is an incredibly flawed take.
So this analysis is a bit off of the mark. From major sins to minor sins.

The problem with this type of historical analysis is that the results are highly dependent on start and end dates. Chose a different set of dates, get a different result. And since historically speaking the future has never been like the past, charts like this really don't convey any real information.

Next, UPRO rebalances daily, you rebalance monthly. So your method has radically lower volatility and thus lower volatility drag.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
alex_686
Posts: 8131
Joined: Mon Feb 09, 2015 2:39 pm

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

Post by alex_686 »

chris319 wrote: Wed Apr 07, 2021 6:28 pm
So you think it is free magic leverage? Come on...
No, I do not think it is "free magic leverage" and never said I did. I corrected the misconception in the post, if you picked up on it.

I wasn't familiar with how swaps work. Thanks for illuminating.
I am not sure how swaps entered the picture. Not sure why they are in the picture. You have to post collateral, so the cost of carry is comparable to margin loans or future contracts.

Also, they act different than a leveraged portfolio using margin or futures contracts. They tend to be total returns, not price swaps. They don't compound. They are relatively illiquid.

It helps to understand that a swap is a collection of forward contracts, which are basically the same as a future contract. Future contracts are a special kind of forward contract.
Last edited by alex_686 on Wed Apr 07, 2021 7:47 pm, edited 1 time in total.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
MotoTrojan
Posts: 11017
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan »

alex_686 wrote: Wed Apr 07, 2021 7:40 pm
MotoTrojan wrote: Wed Apr 07, 2021 6:00 pm
Rogue_trader wrote: Wed Apr 07, 2021 5:54 pm

Volatility drag is totally irrelevant here. I am talking about daily returns and their cumulative product not annual returns! Have you tried to compare UPRO and 3*S&P500? You can easily do it in excel. UPRO beats it and it beats it big.
I think you need to bust out excel a bit more as it apparently isn't as easy as you think. Are you doing 3x returns or 3x the final value? That isn't how compounding works...

3x S&P500 isn't the final wealth gained of S&P500 multiplied by 3x, it is 3x the CAGR, which with linear returns over many years would be WAY more than 3x the wealth, in fact taken to infinity it would be an infinite times more wealth gained.

See here, even with the negative drag of -200% CASHX (so a bit less than 3x S&P500) it is actually UPRO that GETS beat (3x S&P500 has a 43% CAGR while UPRO got a 36.4% CAGR).

https://www.portfoliovisualizer.com/bac ... on3_2=-200

Other than the expense ratio, volatility drag is the ONLY relevant reason why UPRO underperformed 3x S&P500. This is an incredibly flawed take.
So this analysis is a bit off of the mark. From major sins to minor sins.

The problem with this type of historical analysis is that the results are highly dependent on start and end dates. Chose a different set of dates, get a different result. And since historically speaking the future has never been like the past, charts like this really don't convey any real information.

Next, UPRO rebalances daily, you rebalance monthly. So your method has radically lower volatility and thus lower volatility drag.
See below, misunderstood the posters point, they were referring to actual daily returns and suggesting UPRO outperforms 3x the S&P's move.
Last edited by MotoTrojan on Wed Apr 07, 2021 7:56 pm, edited 1 time in total.
Hydromod
Posts: 510
Joined: Tue Mar 26, 2019 10:21 pm

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

Post by Hydromod »

The OP was comparing ^GSPC*3 to UPRO, I think, cumulative daily returns. So price only versus UPRO.
MotoTrojan
Posts: 11017
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan »

Hydromod wrote: Wed Apr 07, 2021 7:49 pm The OP was comparing ^GSPC*3 to UPRO, I think, cumulative daily returns. So price only versus UPRO.
I see, I misunderstood that entirely and apologize to Rogue_trader. I still don't quite understand how that can be the case, that implies that UPRO outperforms its gross index every day? Where is the headwind coming to offset the ER, let alone borrowing costs? I suppose dividends is the answer but Rogue seems to say they are using a total return daily index so not sure how this closes.
chris319
Posts: 262
Joined: Thu Jan 28, 2021 6:04 pm

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

Post by chris319 »

I am not sure how swaps entered the picture. Not sure why they are in the picture.
Index swaps make up a large portion of LETF holdings:

https://finance.yahoo.com/quote/UPRO?p= ... c=fin-srch
GOMEZ: Morticia! Consolidated Fuzz just hit 212 — I sold it for 6 — saved all that tax. MORTICIA: Brilliant!
Hydromod
Posts: 510
Joined: Tue Mar 26, 2019 10:21 pm

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

Post by Hydromod »

I did the comparison between ^GSPC*3 and UPRO from UPRO onset to last Friday. Cumulative product of daily returns.

UPRO ended up about 1.44 times larger than ^GSPC*3 (CAGR 36.4% versus 33.0%). ^GSPC CAGR = 13.4%.

Then I did SPY*3 over the same period. SPY*3 ended up about 1.42 times larger than UPRO (CAGR 41.3% vs. 36.4%). SPY CAGR = 15.7%.

^GSPC is price only I believe.
Rogue_trader
Posts: 19
Joined: Wed Apr 07, 2021 1:14 pm

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

Post by Rogue_trader »

chris319 wrote: Wed Apr 07, 2021 6:12 pm
The same risk free rate should be also emended in UPRO, because UPRO cannot borrow for free, but it seems that it can and not only that they are even borrowing at negative rates. Or there must be some other source of return boost, which makes UPRO to beat 3*S&P500.
What exactly is the fund borrowing? What kind of asset?

We are well aware that these funds promise 2x or 3x the daily return, not annual. Some people don't understand this but we do!
Total return equity swap and possibly some futures...
Rogue_trader
Posts: 19
Joined: Wed Apr 07, 2021 1:14 pm

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

Post by Rogue_trader »

MotoTrojan wrote: Wed Apr 07, 2021 6:00 pm
Rogue_trader wrote: Wed Apr 07, 2021 5:54 pm

Volatility drag is totally irrelevant here. I am talking about daily returns and their cumulative product not annual returns! Have you tried to compare UPRO and 3*S&P500? You can easily do it in excel. UPRO beats it and it beats it big.
I think you need to bust out excel a bit more as it apparently isn't as easy as you think. Are you doing 3x returns or 3x the final value? That isn't how compounding works...

3x S&P500 isn't the final wealth gained of S&P500 multiplied by 3x, it is 3x the CAGR, which with linear returns over many years would be WAY more than 3x the wealth, in fact taken to infinity it would be an infinite times more wealth gained.

See here, even with the negative drag of -200% CASHX (so a bit less than 3x S&P500) it is actually UPRO that GETS beat (3x S&P500 has a 43% CAGR while UPRO got a 36.4% CAGR).

https://www.portfoliovisualizer.com/bac ... on3_2=-200

Other than the expense ratio, volatility drag is the ONLY relevant reason why UPRO underperformed 3x S&P500. This is an incredibly flawed take.
I am all the time doing about daily returns so the volatility drag is not mathematically relevant.
I multiplied every day's return with 3 and then calculated cumulative product. So if S&P500 returns were:
Day1 1%
Day2 2%
S&P500 is then (assuming it starts from 100) 100*(1+1%)*(1+2%)
In case of 3*S&P500 I end up with
Day1 3%
Day2 6%
and then the value is 100*(1+3%)*(1+6%)

So you also see UPRO beating 3x S&P500, right?
Rogue_trader
Posts: 19
Joined: Wed Apr 07, 2021 1:14 pm

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

Post by Rogue_trader »

MotoTrojan wrote: Wed Apr 07, 2021 6:00 pm
Rogue_trader wrote: Wed Apr 07, 2021 5:54 pm

Volatility drag is totally irrelevant here. I am talking about daily returns and their cumulative product not annual returns! Have you tried to compare UPRO and 3*S&P500? You can easily do it in excel. UPRO beats it and it beats it big.
I think you need to bust out excel a bit more as it apparently isn't as easy as you think. Are you doing 3x returns or 3x the final value? That isn't how compounding works...

3x S&P500 isn't the final wealth gained of S&P500 multiplied by 3x, it is 3x the CAGR, which with linear returns over many years would be WAY more than 3x the wealth, in fact taken to infinity it would be an infinite times more wealth gained.

See here, even with the negative drag of -200% CASHX (so a bit less than 3x S&P500) it is actually UPRO that GETS beat (3x S&P500 has a 43% CAGR while UPRO got a 36.4% CAGR).

https://www.portfoliovisualizer.com/bac ... on3_2=-200

Other than the expense ratio, volatility drag is the ONLY relevant reason why UPRO underperformed 3x S&P500. This is an incredibly flawed take.
I am all the time doing about daily returns so the volatility drag is not mathematically relevant.
I multiplied every day's return with 3 and then calculated cumulative product. So if S&P500 returns were:
Day1 1%
Day2 2%
S&P500 is then (assuming it starts from 100) 100*(1+1%)*(1+2%)
In case of 3*S&P500 I end up with
Day1 3%
Day2 6%
and then the value is 100*(1+3%)*(1+6%)

So you also see UPRO beating 3x S&P500, right?
Rogue_trader
Posts: 19
Joined: Wed Apr 07, 2021 1:14 pm

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

Post by Rogue_trader »

MotoTrojan wrote: Wed Apr 07, 2021 7:46 pm
alex_686 wrote: Wed Apr 07, 2021 7:40 pm
MotoTrojan wrote: Wed Apr 07, 2021 6:00 pm
Rogue_trader wrote: Wed Apr 07, 2021 5:54 pm

Volatility drag is totally irrelevant here. I am talking about daily returns and their cumulative product not annual returns! Have you tried to compare UPRO and 3*S&P500? You can easily do it in excel. UPRO beats it and it beats it big.
I think you need to bust out excel a bit more as it apparently isn't as easy as you think. Are you doing 3x returns or 3x the final value? That isn't how compounding works...

3x S&P500 isn't the final wealth gained of S&P500 multiplied by 3x, it is 3x the CAGR, which with linear returns over many years would be WAY more than 3x the wealth, in fact taken to infinity it would be an infinite times more wealth gained.

See here, even with the negative drag of -200% CASHX (so a bit less than 3x S&P500) it is actually UPRO that GETS beat (3x S&P500 has a 43% CAGR while UPRO got a 36.4% CAGR).

https://www.portfoliovisualizer.com/bac ... on3_2=-200

Other than the expense ratio, volatility drag is the ONLY relevant reason why UPRO underperformed 3x S&P500. This is an incredibly flawed take.
So this analysis is a bit off of the mark. From major sins to minor sins.

The problem with this type of historical analysis is that the results are highly dependent on start and end dates. Chose a different set of dates, get a different result. And since historically speaking the future has never been like the past, charts like this really don't convey any real information.

Next, UPRO rebalances daily, you rebalance monthly. So your method has radically lower volatility and thus lower volatility drag.
See below, misunderstood the posters point, they were referring to actual daily returns and suggesting UPRO outperforms 3x the S&P's move.
Volatility drag plays no role, I am looking at daily returns only.
Rogue_trader
Posts: 19
Joined: Wed Apr 07, 2021 1:14 pm

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

Post by Rogue_trader »

MotoTrojan wrote: Wed Apr 07, 2021 6:16 pm
chris319 wrote: Wed Apr 07, 2021 6:12 pm
The same risk free rate should be also emended in UPRO, because UPRO cannot borrow for free, but it seems that it can and not only that they are even borrowing at negative rates. Or there must be some other source of return boost, which makes UPRO to beat 3*S&P500.
What exactly is the fund borrowing? What kind of asset?

We are well aware that these funds promise 2x or 3x the daily return, not annual. Some people don't understand this but we do!
So you think it is free magic leverage? Come on...

It is called a swap.

https://www.investopedia.com/terms/s/sw ... in%20swaps.
Total Return Swaps
In a total return swap, the total return from an asset is exchanged for a fixed interest rate. This gives the party paying the fixed-rate exposure to the underlying asset—a stock or an index. For example, an investor could pay a fixed rate to one party in return for the capital appreciation plus dividend payments of a pool of stocks.
You say "we" understand, but the post you quoted is saying an outright lie that UPRO outperformed 3x S&P500.
I of course know UPRO should be lagging because of the expense ratio and cost of borrowing from the floating leg of the swap, but that is not what I see and I have really no idea why. It makes no sense at all.
Rogue_trader
Posts: 19
Joined: Wed Apr 07, 2021 1:14 pm

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

Post by Rogue_trader »

Hydromod wrote: Wed Apr 07, 2021 7:49 pm The OP was comparing ^GSPC*3 to UPRO, I think, cumulative daily returns. So price only versus UPRO.
^GSPC is total return index, isn't it? Look at the comparison with SPY.
https://finance.yahoo.com/chart/%5EGSPC ... QifX19fQ--
Rogue_trader
Posts: 19
Joined: Wed Apr 07, 2021 1:14 pm

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

Post by Rogue_trader »

Hydromod wrote: Wed Apr 07, 2021 9:53 pm I did the comparison between ^GSPC*3 and UPRO from UPRO onset to last Friday. Cumulative product of daily returns.

UPRO ended up about 1.44 times larger than ^GSPC*3 (CAGR 36.4% versus 33.0%). ^GSPC CAGR = 13.4%.

Then I did SPY*3 over the same period. SPY*3 ended up about 1.42 times larger than UPRO (CAGR 41.3% vs. 36.4%). SPY CAGR = 15.7%.

^GSPC is price only I believe.
Is ^GSPC really price only? It fits SPY perfectly.
Look at this
https://finance.yahoo.com/chart/%5EGSPC ... VrIn19fQ--
Last edited by Rogue_trader on Thu Apr 08, 2021 3:07 am, edited 1 time in total.
AnilG
Posts: 68
Joined: Wed Sep 09, 2015 1:41 am

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

Post by AnilG »

After following several LETF discussion threads on Bogleheads for last couple of years, I observed few things.
  • Analysis paralysis is real. Most discussion get into the weeds losing sight of the goal.
  • No one reads original two threads completely. Over the years, I have noticed same discussion of trying different combinations instead of UPRO/TMF and TQQQ/TMF, “this time is different”, interest rate environment is changing, etc at least three times with no conclusion and/or different result/outcome.
  • Still no one has come up with anything better or cohesive or a solid counter argument than what HedgeFundie came up with originally.
jarjarM
Posts: 1034
Joined: Mon Jul 16, 2018 1:21 pm

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

Post by jarjarM »

Rogue_trader wrote: Thu Apr 08, 2021 2:13 am
Hydromod wrote: Wed Apr 07, 2021 9:53 pm I did the comparison between ^GSPC*3 and UPRO from UPRO onset to last Friday. Cumulative product of daily returns.

UPRO ended up about 1.44 times larger than ^GSPC*3 (CAGR 36.4% versus 33.0%). ^GSPC CAGR = 13.4%.

Then I did SPY*3 over the same period. SPY*3 ended up about 1.42 times larger than UPRO (CAGR 41.3% vs. 36.4%). SPY CAGR = 15.7%.

^GSPC is price only I believe.
Is ^GSPC really price only? It fits SPY perfectly.
Look at this
https://finance.yahoo.com/chart/%5EGSPC ... VrIn19fQ--
^GSPC is price index only. If your data source is yahoo, there’s 2 column in the historical dataset, close is price index only, adjClose is total return. ^GSPC close matches exactly as adjClose.
Post Reply