Why are leveraged funds (2x/3x) so expensive?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Topic Author
psteinx
Posts: 5785
Joined: Tue Mar 13, 2007 2:24 pm

Why are leveraged funds (2x/3x) so expensive?

Post by psteinx »

Many, many ETFs are available at ERs of < 0.10%. VOO is, I think, 0.03%.

But leveraged ETFs all seem to be MUCH more expensive. UPRO, which many like for 3x S&P 500 leverage, is 0.93%.

That's a high # in and of itself. But even if you adjust for the leverage, it's still high - it's loosely like stacking 3 VOOs. So, for the extra 2 legs (the leveraged part), you're paying about 0.45% overall, or 0.42% more than VOO.

I *think* UPRO is fairly big, so it's shouldn't really be a scale issue. And, IIUC, what they're doing, either trading futures directly, or perhaps negotiating swap agreements with Goldman Sachs or whoever, shouldn't be THAT hard to do.

IIUC, the investor is still, in some fashion (presumably implicit in the futures and/or swap contracts) paying interest on the leverage, in addition to the ER.

So, why hasn't price competition driven down ERs of leveraged funds?

My suspicion is that there's some residual exposure in tail events (much moreso than for VOO), for the fund provider, and that, as a result, providers aren't willing to go TOO low on the ERs. But that's just a guess. I'm open to other thoughts/information.
Last edited by psteinx on Fri Sep 17, 2021 3:45 pm, edited 1 time in total.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Why are leveraged funds (2x/3x) so expensive?

Post by Marseille07 »

I don't see much competition. SPXL is 1.01%, SSO is 0.91%. Sophisticated traders can use futures / swaps, but most plebs prefer simply longing the ETFs.
Topic Author
psteinx
Posts: 5785
Joined: Tue Mar 13, 2007 2:24 pm

Re: Why are leveraged funds (2x/3x) so expensive?

Post by psteinx »

Sophistication is likely part of it, but there are other reasons to potentially prefer a 3X ETF, including taxes.
User avatar
dziuniek
Posts: 1402
Joined: Mon Jul 23, 2012 2:54 pm

Re: Why are leveraged funds (2x/3x) so expensive?

Post by dziuniek »

For 3X ETFs, SEC said no new 3X ETFs can be created. So there will be no new competition atleast for now.

Doesn't seem that expensive when you look at alternatives like margin.
Get rich or die tryin'
jarjarM
Posts: 2502
Joined: Mon Jul 16, 2018 1:21 pm

Re: Why are leveraged funds (2x/3x) so expensive?

Post by jarjarM »

dziuniek wrote: Fri Sep 17, 2021 3:47 pm For 3X ETFs, SEC said no new 3X ETFs can be created. So there will be no new competition atleast for now.
+1, no new competition to push ER down.
User avatar
nisiprius
Advisory Board
Posts: 52105
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Why are leveraged funds (2x/3x) so expensive?

Post by nisiprius »

Something resembling a loan needs to be negotiated every day. And the daily rebalancing seemingly means that stocks need to be bought or sold every day. Maybe those expenses add up.

But the days when you could assume that ETFs were low cost seem to be long gone.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
gougou
Posts: 1311
Joined: Thu Sep 28, 2017 7:42 pm

Re: Why are leveraged funds (2x/3x) so expensive?

Post by gougou »

They need to do a lot of rebalancing daily. Say it's a 3x ETF with $300M stocks, $200M debt so $100M NAV. If market moves up 2% in a day, they have $306M stocks, $200M debt, $106M NAV. To maintain 3x leverage they need to reach $106M x 3 = $318M stocks, so they must immediately borrow $12M and buy $12M of stocks, which is 12% of $100M NAV just in a single day. If they lose 0.1% to trading costs on those 12% transactions that's 0.012% per day and about 3% annualized.

If market moves down 2% next day they need to immediately sell like $12M of stock. They just keep losing money to commissions and bid-ask spread everyday the market moves.
The sillier the market’s behavior, the greater the opportunity for the business like investor.
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: Why are leveraged funds (2x/3x) so expensive?

Post by skierincolorado »

The fees aren't that high. The fee on UPRO is only 3 times higher than the very very low fee on SPY and lower than a lot of others.

But wait! Isn't the fee on UPRO 9x higher (0.9% vs 0.1%)? Great question!

If I have 100k in cash and I want to buy 100k of the S&P500 I can either:

1) Buy 100k of SPY and pay 0.1% of 100k = $100

2) Buy 33.3k of UPRO, park 66.7k in MMA*, and pay 0.9% of 33k = $300

The fee is 3x higher - not 9x.

The fee is likely 3x higher to account for the transaction costs of daily rebalancing and because it is a less competitive market. A 0.3% is on the higher side, but not terribly bad for the privelege of completely hands free 3x leveraged investing.

*note the 66.7k in MMA is not profit - this should very nearly exactly compensate you for the borrowing cost incurred by UPRO who borrowed 66.7k on your behalf at rates very nearly identical to what you'll earn in the MMA. Thus, other than fees, positons #1 and #2 are identical.
Last edited by skierincolorado on Fri Sep 17, 2021 5:01 pm, edited 3 times in total.
tomsense76
Posts: 1428
Joined: Wed Oct 14, 2020 1:52 am

Re: Why are leveraged funds (2x/3x) so expensive?

Post by tomsense76 »

skierincolorado wrote: Fri Sep 17, 2021 4:48 pm The fees aren't that high. The fee on UPRO is only 3 times higher than the very very low fee on SPY and lower than a lot of others.

But wait! Isn't the fee on UPRO 9x higher (0.9% vs 0.1%)? Great question!

If I have 100k in cash and I want to buy 100k of the S&P500 I can either:

1) Buy 100k of SPY and pay 0.1% of 100k = $100

2) Buy 33.3k of UPRO, park 66.7k in MMA, and pay 0.9% of 33k = $300

The fee is 3x higher - not 9x.

The fee is likely 3x higher to account for the transaction costs of daily rebalancing and because it is a less competitive market. A 0.3% is on the higher side, but not terribly bad for the privelege of completely hands free 3x leveraged investing.
Thank you for sharing this analysis! Just goes to show looking at a percentage or an ER is not sufficient to understand the true cost of ownership.

Have seen other variations of this error like preferring LTCG to STCG even though recently purchased shares might not have much gain to speak of. So would have lower overall tax even though they would be STCG.
"Anyone who claims to understand quantum theory is either lying or crazy" -- Richard Feynman
alex_686
Posts: 13286
Joined: Mon Feb 09, 2015 1:39 pm

Re: Why are leveraged funds (2x/3x) so expensive?

Post by alex_686 »

nisiprius wrote: Fri Sep 17, 2021 4:28 pm Something resembling a loan needs to be negotiated every day. And the daily rebalancing seemingly means that stocks need to be bought or sold every day. Maybe those expenses add up.
They use futures. So no negotiation on margin loans or things like that. If they did use margin it would be counted as a expense. But not the implied cost of carry for futures.
gougou wrote: Fri Sep 17, 2021 4:48 pm If market moves down 2% next day they need to immediately sell like $12M of stock. They just keep losing money to commissions and bid-ask spread everyday the market moves.
Trading costs are not part of the expense ratio. So while that is a concern it does not explain OP question. Besides, future contracts are very liquid and have rock bottom trading costs - both explicit and implicit.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
alex_686
Posts: 13286
Joined: Mon Feb 09, 2015 1:39 pm

Re: Why are leveraged funds (2x/3x) so expensive?

Post by alex_686 »

skierincolorado wrote: Fri Sep 17, 2021 4:48 pm The fee is 3x higher - not 9x.
I don't think your argument holds water. I can see where you intuition is going. However, as somebody who has actually calculated a mutual fund expense ratio I don't see a clear linkage leverage and the expense ratio.

They don't use margin. Trading expenses are not counted in the expense ratio.

The expense ratio goes towards running the fund (annual reports and such) and the portfolio manager. Even for a 3x leverage product what they are doing is pretty plain jane stuff. Its not like the accounting becomes 3x as complicated or you need to hire 3x the traders.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
tomsense76
Posts: 1428
Joined: Wed Oct 14, 2020 1:52 am

Re: Why are leveraged funds (2x/3x) so expensive?

Post by tomsense76 »

alex_686 wrote: Fri Sep 17, 2021 5:07 pm
skierincolorado wrote: Fri Sep 17, 2021 4:48 pm The fee is 3x higher - not 9x.
I don't think your argument holds water. I can see where you intuition is going. However, as somebody who has actually calculated a mutual fund expense ratio I don't see a clear linkage leverage and the expense ratio.

They don't use margin. Trading expenses are not counted in the expense ratio.

The expense ratio goes towards running the fund (annual reports and such) and the portfolio manager. Even for a 3x leverage product what they are doing is pretty plain jane stuff. Its not like the accounting becomes 3x as complicated or you need to hire 3x the traders.
Interesting. So the daily rebalancing doesn't factor into this?

What would you expect the true expense to run this fund to be? The same as SPY? (1.y) times SPY's ER?
"Anyone who claims to understand quantum theory is either lying or crazy" -- Richard Feynman
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: Why are leveraged funds (2x/3x) so expensive?

Post by skierincolorado »

alex_686 wrote: Fri Sep 17, 2021 5:07 pm
skierincolorado wrote: Fri Sep 17, 2021 4:48 pm The fee is 3x higher - not 9x.
I don't think your argument holds water. I can see where you intuition is going. However, as somebody who has actually calculated a mutual fund expense ratio I don't see a clear linkage leverage and the expense ratio.

They don't use margin. Trading expenses are not counted in the expense ratio.

The expense ratio goes towards running the fund (annual reports and such) and the portfolio manager. Even for a 3x leverage product what they are doing is pretty plain jane stuff. Its not like the accounting becomes 3x as complicated or you need to hire 3x the traders.
Oh I agree they are making lots more money than SPY is (if they had the same AUM).

But I'm not looking at it from the fund's perspective. From an investor's perspective, the fee is 3x higher.

Say I had 100k in my account and I want to own 300k of S&P500. I have two choices using ETFs (ignoring futures):

1) Take out a margin loan for 200k and buy 300k of SPY. I pay $300 in fees on 300k of SPY.

2) Buy 100k of UPRO. I pay $900 of fees on UPRO.

The fee is 3x higher in #2 than #1. Not 9x. The borrowing costs are also much lower in #2 than in #1, because UPRO borrows near LIBOR and I borrow at whatever crappy rate my broker has, at best 1% at IBKR.

So from an investor perspective the fee is 3x higher. But there are several benefits to this higher fee, such as hands off leverage and low borrowing costs. This creates demand for the fund and explains why investors are willing to pay the 3x fee. And if you are correct that the cost of operating the fund are the same as SPY, the fund is making 9x more money. However, I suspect that their are costs are fairly proportional with leverage. When SPY receives 1M in cash, they buy 1M in stock, and incur fees on those transactions. When UPRO receives 1M in cash, they buy 3M of stock (or derivates) and incur operating expenses on 3M of transactions. Thus the amount of transactions are proportional to the leverage ratio. So their costs are 3x higher, while the fee is 9x higher in absolute terms. So they are probably doing pretty well - except the AUM is pretty low so there is no economy of scale. The lack of economy of scale is probably the bigger factor here and makes there expenses pretty high. On the other hand it's not a competitive market, and there is strong demand for these funds due to the aforementioned benefits (hands off leverage, low borrowing costs), so they can charge a higher fee.

It's two separate questions really. How much more is it costing the investor to gain exposure to the S&P500? How much are the funds actual expenses relative to AUM? You're answering #2, but I'm really trying to answer #1 and speculating a little about #2.

And again, from an investor perspective, what one cares about is the fee per unit of exposure to the market you are getting.
Last edited by skierincolorado on Fri Sep 17, 2021 5:36 pm, edited 1 time in total.
alex_686
Posts: 13286
Joined: Mon Feb 09, 2015 1:39 pm

Re: Why are leveraged funds (2x/3x) so expensive?

Post by alex_686 »

tomsense76 wrote: Fri Sep 17, 2021 5:22 pm Interesting. So the daily rebalancing doesn't factor into this?

What would you expect the true expense to run this fund to be? The same as SPY? (1.y) times SPY's ER?
The daily rebalance wouldn't. All funds have to trade every day. Maybe due to the complexity it might add an extra 10% or so. Maybe. This is a pretty dead simple fund to run. It is not like they have 500 stocks to rebalance every day. IIRC they just have 2 or 3 different futures that they trade.

SPY has some advantages. Size - the bigger it is the more it can spread around the costs. It can lend out securities and generate income that way.

I like the theory that it has the market to itself so is jacking up the expense ratio to make a nice fat profit. I would want to take a look at the annual report and compare it with a few others from the same fund family to confirm. But it is a nice solid theory.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: Why are leveraged funds (2x/3x) so expensive?

Post by skierincolorado »

alex_686 wrote: Fri Sep 17, 2021 5:35 pm
tomsense76 wrote: Fri Sep 17, 2021 5:22 pm Interesting. So the daily rebalancing doesn't factor into this?

What would you expect the true expense to run this fund to be? The same as SPY? (1.y) times SPY's ER?
The daily rebalance wouldn't. All funds have to trade every day. Maybe due to the complexity it might add an extra 10% or so. Maybe. This is a pretty dead simple fund to run. It is not like they have 500 stocks to rebalance every day. IIRC they just have 2 or 3 different futures that they trade.

SPY has some advantages. Size - the bigger it is the more it can spread around the costs. It can lend out securities and generate income that way.

I like the theory that it has the market to itself so is jacking up the expense ratio to make a nice fat profit. I would want to take a look at the annual report and compare it with a few others from the same fund family to confirm. But it is a nice solid theory.
I don't think that's quite right. Transactions have costs associated with them. Making more transactions because you have to daily rebalance, and because you have to make 3x more transactions because of the leverage ratio, surely has a cost associated with it. I'm sure these funds make money, but I doubt they are earning some absurd profit margin like 90%.
Last edited by skierincolorado on Fri Sep 17, 2021 5:45 pm, edited 1 time in total.
Topic Author
psteinx
Posts: 5785
Joined: Tue Mar 13, 2007 2:24 pm

Re: Why are leveraged funds (2x/3x) so expensive?

Post by psteinx »

The SPY ER is terrible. Trying to use it as a comparison for best case alternative is bad analysis.

As another has posted - the commissions that UPRO pays don't go to ER.

Yes, assuming they use futures, they must do some trading each day to rebalance. But, assuming futures markets are difficult, that's not a big ask of an ETF.

The SEC limiting competition on 3x funds - I can see how that would reduce ER pressure. Does that apply to 2x funds as well?
Topic Author
psteinx
Posts: 5785
Joined: Tue Mar 13, 2007 2:24 pm

Re: Why are leveraged funds (2x/3x) so expensive?

Post by psteinx »

skierincolorado wrote: Fri Sep 17, 2021 5:42 pmThe daily rebalancing might not incur much in costs, but the leverage ratio certainly must. It's more expensive to make 3M of transactions than 1M.
IIUC, typical trading-related hard expenses - crossing the bid/ask, and actual commissions (and, presumably, exchange/SEC fees) are normally borne by a funds' investors, so they don't explain a high ER. Yes, there needs to be human beings (or maybe really good algorithms) executing those trades (which, presumably DOES come at the fund provider's expense), but in a liquid market, this trading shouldn't be THAT hard...
alex_686
Posts: 13286
Joined: Mon Feb 09, 2015 1:39 pm

Re: Why are leveraged funds (2x/3x) so expensive?

Post by alex_686 »

skierincolorado wrote: Fri Sep 17, 2021 5:42 pm The daily rebalancing might not incur much in costs, but the leverage ratio certainly must. It's more expensive to make 3M of transactions than 1M.
First, futures are very liquid. It does not take that much effort to enter in a order for 100m verse 300m. Now, IIRC, back 10 years ago the 3x leveraged products was causing the market prices to go wonky due to their heavy trading. But this is not counted as a expense.

Second, trading expenses don't make it into the expense ratio.

Third - how would leverage increase expenses? I can see maybe a little bit around the edges? But I can't think of any expense or accounting mechanism that would have a major impact.

So no, I really can't see how the leverage ratio would matter.
Last edited by alex_686 on Fri Sep 17, 2021 5:51 pm, edited 1 time in total.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: Why are leveraged funds (2x/3x) so expensive?

Post by skierincolorado »

psteinx wrote: Fri Sep 17, 2021 5:45 pm
skierincolorado wrote: Fri Sep 17, 2021 5:42 pmThe daily rebalancing might not incur much in costs, but the leverage ratio certainly must. It's more expensive to make 3M of transactions than 1M.
IIUC, typical trading-related hard expenses - crossing the bid/ask, and actual commissions (and, presumably, exchange/SEC fees) are normally borne by a funds' investors, so they don't explain a high ER. Yes, there needs to be human beings (or maybe really good algorithms) executing those trades (which, presumably DOES come at the fund provider's expense), but in a liquid market, this trading shouldn't be THAT hard...
Yes I'm talking about the transaction costs incurred by the fund itself.

Anyways, all that really matters is from an investor perspective.

To gain exposure to 100k of S&P500 will cost you:

1) UPRO: $300

2) SPY: $100

3) lowest cost funds like the VTI or Fidelity zero fee funds: $0-$50

So it's $300 to get 100k of exposure, vs $0 in the best case. I guess the point isn't that it's 3x higher, it's that it's .3% - not .9%.

Paying .3% for the benefit of low borrowing cost and hands off leverage is worth it to a lot of people who are seeking either of those. The alternatives (margin) can be much more expensive.
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: Why are leveraged funds (2x/3x) so expensive?

Post by skierincolorado »

alex_686 wrote: Fri Sep 17, 2021 5:48 pm
skierincolorado wrote: Fri Sep 17, 2021 5:42 pm The daily rebalancing might not incur much in costs, but the leverage ratio certainly must. It's more expensive to make 3M of transactions than 1M.
First, futures are very liquid. It does not take that much effort to enter in a order for 100m verse 300m. Now, IIRC, back 10 years ago the 3x leveraged products was causing the market prices to go wonky due to their heavy trading. But this is not counted as a expense.

Second, trading expenses don't make it into the expense ratio.

So no, I really can't see how the leverage ratio would matter.
I can see your point, you might be right. It might come down more to the lack of economy of scale then.

Either way, from an investor perspective, the fee is really .3% to get the same exposure as SPY. And .3% isn't that high, when the alternative (margin) is a lot more expensive and time consuming.

If UPRO only charged anything less than .3% it would be cheaper than SPY. One could simply own 1/3 as much UPRO and keep cash in a MMA and rebalance. Assuming you had market rates on your MMA, this would be an identical position with lower fees.

Of course the real benefit is keeping that extra cash in something like intermediate term treasuries.
Last edited by skierincolorado on Fri Sep 17, 2021 5:55 pm, edited 1 time in total.
Topic Author
psteinx
Posts: 5785
Joined: Tue Mar 13, 2007 2:24 pm

Re: Why are leveraged funds (2x/3x) so expensive?

Post by psteinx »

skierincolorado wrote: Fri Sep 17, 2021 5:51 pm Paying .3% for the benefit of low borrowing cost and hands off leverage is worth it to a lot of people who are seeking either of those. The alternatives (margin) can be much more expensive.
It's not .3%.

The alternative is not SPY (a bad fund for a buy and holder), it's VOO*, at 0.03%. 3x 0.03% is 0.09%. UPRO is 0.93%, so that leverage (going from 100% to 300%) costs ER on each X of the levered part of 0.45%, or 0.42% above VOO.

Now, UPRO may still be a useful fund for an investor anyways, but it's disappointing to me if that's best of breed, in a relatively mature, large space like levered ETFs.

* Actually, I like VTI even better, but I would assume that would be hard to do levered at the fund level in the manner of UPRO.
Last edited by psteinx on Fri Sep 17, 2021 5:57 pm, edited 1 time in total.
User avatar
typical.investor
Posts: 5245
Joined: Mon Jun 11, 2018 3:17 am

Re: Why are leveraged funds (2x/3x) so expensive?

Post by typical.investor »

psteinx wrote: Fri Sep 17, 2021 5:42 pm The SEC limiting competition on 3x funds - I can see how that would reduce ER pressure.
Sigh, and yet there are completely unregulated securities markets much larger than 3x funds operating. Why do they have to pick on 3X funds? Because people need to know that the only way to offset volatility drag is by taking money out when it does well and putting more in when it doesn’t? That’s the only way you can earn 3X (minus costs).

** the unregulated securities markets are not allowed to be discussed here. Suffice it to say there is a huge, active market for things which very much look to meet the definition of securities in the 1933 Securities Act.
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: Why are leveraged funds (2x/3x) so expensive?

Post by skierincolorado »

psteinx wrote: Fri Sep 17, 2021 5:55 pm
skierincolorado wrote: Fri Sep 17, 2021 5:51 pm Paying .3% for the benefit of low borrowing cost and hands off leverage is worth it to a lot of people who are seeking either of those. The alternatives (margin) can be much more expensive.
It's not .3%.

The alternative is not SPY (a bad fund for a buy and holder), it's VOO, at 0.03%. 3x 0.03% is 0.09%. UPRO is 0.93%, so that leverage (going from 100% to 300%) costs ER on each X of the levered part of 0.45%, or 0.42% above VOO.

Now, UPRO may still be a useful fund for an investor anyways, but it's disappointing to me if that's best of breed, in a relativel mature, large space like levered ETFs.
It is .3% per unit of S&P500. You are correct that it's more than 3x VOO, but it's still .3%. VOO is .03% per unit of S&P500. UPRO is .3% per unit of S&P500, so it's 10x as expensive as VOO, and 3x as expensive as SPY - per unit of S&P500.

The 10x is a little misleading because it's basically 10x nothing. If and when a 0% fee S&P500 fund comes out how will we compare then? Probably best to look at it in absolute terms. Per unit of S&P500 you pay:

.03% for VOO
.1% for SPY
.31% for UPRO

The .31% fee is going to be considered quite reasonable by many given how much time and borrowing costs you can save by avoiding broker margin.
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: Why are leveraged funds (2x/3x) so expensive?

Post by skierincolorado »

typical.investor wrote: Fri Sep 17, 2021 5:56 pm
Because people need to know that the only way to offset volatility drag is by taking money out when it does well and putting more in when it doesn’t? That’s the only way you can earn 3X (minus costs).
Not really true. In theory, volatility drag shouldn't exist in the long-run. If volatility drag was a real, permanent feature, you could make guaranteed money by shorting UPRO and buying SPY in proportion. But it can exist in volatile markets and the market can remain irrational longer than you can wait, so I agree it's best to try to avoid. The market can also be irrationally low-volatility and during those times you'll experience the opposite of volatility drag... you'll get returns more than the gearing ratio would suggest. UPRO for example is 32x since inception, while SPY is just 5x. Per the gearing ratio UPRO should only be 15x. But it's 30x - the opposite of volatility decay. Yes the markets been very up, but some big drops too.

SSO is getting pretty close to 2x SPY's return since 2007. If the market keeps going up with low volatility, SSO will surpass 2x SPY returns over the course of its life. And that's after fees and borrowing costs! It's probably already surpassed 2x SPY returns pre-fees pre-borrowing costs.

People like to reference ULPIX since the late 1990s - but ULPIX had 1.7% fees and who knows how well they implemented their financing. Did futures even exist at their inception?
User avatar
typical.investor
Posts: 5245
Joined: Mon Jun 11, 2018 3:17 am

Re: Why are leveraged funds (2x/3x) so expensive?

Post by typical.investor »

skierincolorado wrote: Fri Sep 17, 2021 6:04 pm
typical.investor wrote: Fri Sep 17, 2021 5:56 pm
Because people need to know that the only way to offset volatility drag is by taking money out when it does well and putting more in when it doesn’t? That’s the only way you can earn 3X (minus costs).
Not really true. In theory, volatility drag shouldn't exist in the long-run. If volatility drag was a real, permanent feature, you could make guaranteed money by shorting UPRO and buying SPY in proportion. But it can exist in volatile markets and the market can remain irrational longer than you can wait, so I agree it's best to try to avoid.

SSO is getting pretty close to 2x SPY's return since 2007. If the market keeps going up with low volatility, SSO will surpass 2x SPY returns over the course of its life. And that's after fees and borrowing costs! It's probably already surpassed 2x SPY returns pre-fees pre-borrowing costs.

People like to reference ULPIX since the late 1990s - but ULPIX had 1.7% fees and who knows how well they implemented their financing.
If the assertion is that a 3X fund should return 3X the index in the long term because volatility has no effect, then I would suggest you to learn about it a little more. Just sorry, but no.

This has nothing to do with the market being irrational. It simply has to do with how returns correspond to price movement when leverage is reset daily.

I rather contest the assertion that the volatility effect is not a fixed permanent condition. Simply stated, one can not short UPRO and go long SPY in proportion and be guaranteed anything.

For example, shorting UPRO and with 25% volatility when the index shows a 20% loss will lower your losses compared to 3X SPY. Similarly, shorting shorting UPRO and with 25% volatility when the index shows a 40% gain will increase your profits compared to 3X SPY.

In fact, I'd say that shorting UPRO and going long 3X spy helps you the most when volatility and returns are both high and hurts you the most when volatility is low and returns are either very good or very bad. In any case, with volatility often running between 10-25%, whether shorting UPRO and going long 3x SPY helps you or hurts you really depends on the returns and the sequence of returns and volatility on a daily basis.

If you do the math, you will see what I said is correct. The only way for 3X leveraged funds to return 3X the index is by adding money when it does poorly and removing when it does well. You can do this on a daily basis but you have to contend with a cash drag as you keep a pile of cash available. HFEA does it quarterly using 3X LTT.

For the strategy, if UPRO and TMF both need cash infusions the same time (i.e. stagflation), you might think about adding extra cash so you don't suffer a permanent loss.

I am really sorry, but I dismiss your contention as being factually incorrect as shown in 3X ETFs prospectus.

Image
Last edited by typical.investor on Fri Sep 17, 2021 7:01 pm, edited 1 time in total.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Why are leveraged funds (2x/3x) so expensive?

Post by Marseille07 »

typical.investor wrote: Fri Sep 17, 2021 5:56 pm
psteinx wrote: Fri Sep 17, 2021 5:42 pm The SEC limiting competition on 3x funds - I can see how that would reduce ER pressure.
Sigh, and yet there are completely unregulated securities markets much larger than 3x funds operating. Why do they have to pick on 3X funds? Because people need to know that the only way to offset volatility drag is by taking money out when it does well and putting more in when it doesn’t? That’s the only way you can earn 3X (minus costs).

** the unregulated securities markets are not allowed to be discussed here. Suffice it to say there is a huge, active market for things which very much look to meet the definition of securities in the 1933 Securities Act.
To be fair, the new SEC chair Gensler is going after the unregulated securities markets as well.
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: Why are leveraged funds (2x/3x) so expensive?

Post by skierincolorado »

typical.investor wrote: Fri Sep 17, 2021 6:54 pm
skierincolorado wrote: Fri Sep 17, 2021 6:04 pm
typical.investor wrote: Fri Sep 17, 2021 5:56 pm
Because people need to know that the only way to offset volatility drag is by taking money out when it does well and putting more in when it doesn’t? That’s the only way you can earn 3X (minus costs).
Not really true. In theory, volatility drag shouldn't exist in the long-run. If volatility drag was a real, permanent feature, you could make guaranteed money by shorting UPRO and buying SPY in proportion. But it can exist in volatile markets and the market can remain irrational longer than you can wait, so I agree it's best to try to avoid.

SSO is getting pretty close to 2x SPY's return since 2007. If the market keeps going up with low volatility, SSO will surpass 2x SPY returns over the course of its life. And that's after fees and borrowing costs! It's probably already surpassed 2x SPY returns pre-fees pre-borrowing costs.

People like to reference ULPIX since the late 1990s - but ULPIX had 1.7% fees and who knows how well they implemented their financing.
If the assertion is that a 3X fund should return 3X the index in the long term because volatility has no effect, then I would suggest you to learn about it a little more. Just sorry, but no.

This has nothing to do with the market being irrational. It simply has to do with how returns correspond to price movement when leverage is reset daily.

I rather contest the assertion that the volatility effect is not a fixed permanent condition. Simply stated, one can not short UPRO and go long SPY in proportion and be guaranteed anything.

For example, shorting UPRO and with 25% volatility when the index shows a 20% loss will lower your losses compared to 3X SPY. Similarly, shorting shorting UPRO and with 25% volatility when the index shows a 40% gain will increase your profits compared to 3X SPY.

In fact, I'd say that shorting UPRO and going long 3X spy helps you the most when volatility and returns are both high and hurts you the most when volatility is low and returns are either very good or very bad. In any case, with volatility often running between 10-25%, whether shorting UPRO and going long 3x SPY helps you or hurts you really depends on the returns and the sequence of returns and volatility on a daily basis.

If you do the math, you will see what I said is correct. The only way for 3X leveraged funds to return 3X the index is by adding money when it does poorly and removing when it does well. You can do this on a daily basis but you have to contend with a cash drag as you keep a pile of cash available. HFEA does it quarterly using 3X LTT.

For the strategy, if UPRO and TMF both need cash infusions the same time (i.e. stagflation), you might think about adding extra cash so you don't suffer a permanent loss.

I am really sorry, but I dismiss your contention and being factually incorrect as shown in 3X ETFs prospectus.
I've done extensive research on the subject and I think you are not properly thinking through how shorting UPRO and buying 3x SPY would be a guaranteed risk-free return if what you are saying is true.

UPRO has returned 30x since inception while SPY has returned 5x. That's 6x better not 3x. No volatility decay there, quite the opposite. Yes the market is mostly up, but there are ups and downs every day, and some big drops in there along the way too.

SSO has returned nearly 2x SPY since inception in 2007. That's despite fees, borrowing costs, and a 50% market crash. On a pre-fee pre-borrowing cost measure, it's likely over 2x SPY - no volatility decay here either.

The classic example is ULPIX, but ULPIX had 1.7% fees, borrowing costs were higher in the 1990s, and we don't know how efficient its execution was.

The simba spreadsheet simulates 3x returns since 1955. I'm not sure how frequently they rebalanced - it might be annual, but if volatility decay is a real thing it should be evident in both daily and annual rebalancing. It's not. The 3x simulated return since 1955 is very nearly 3x the S&P500 return, especially after considering borrowing costs.

You're simply incorrect. If volatility decay were a real theoretical construct, then it could be aribtraged and eliminated by hedge funds.

The statement is in the prospectus because returns of intermediate periods can deviate significantly from the gearing ratio. But over long enough periods, in an efficient market, theoretically the returns should closely match the gearing ratio minus financing costs and fees. The financing costs and fees are significant and could be mistaken as volatility decay. The strongest evidence of this is probably SSO since 2007 which has returned 1.65x SPY. This is despite fees, borrowing costs, and a 50%+ market crash in 2008 and a 30% market crash in 2020.
User avatar
typical.investor
Posts: 5245
Joined: Mon Jun 11, 2018 3:17 am

Re: Why are leveraged funds (2x/3x) so expensive?

Post by typical.investor »

skierincolorado wrote: Fri Sep 17, 2021 7:02 pm
typical.investor wrote: Fri Sep 17, 2021 6:54 pm
skierincolorado wrote: Fri Sep 17, 2021 6:04 pm
typical.investor wrote: Fri Sep 17, 2021 5:56 pm
Because people need to know that the only way to offset volatility drag is by taking money out when it does well and putting more in when it doesn’t? That’s the only way you can earn 3X (minus costs).
Not really true. In theory, volatility drag shouldn't exist in the long-run. If volatility drag was a real, permanent feature, you could make guaranteed money by shorting UPRO and buying SPY in proportion. But it can exist in volatile markets and the market can remain irrational longer than you can wait, so I agree it's best to try to avoid.

SSO is getting pretty close to 2x SPY's return since 2007. If the market keeps going up with low volatility, SSO will surpass 2x SPY returns over the course of its life. And that's after fees and borrowing costs! It's probably already surpassed 2x SPY returns pre-fees pre-borrowing costs.

People like to reference ULPIX since the late 1990s - but ULPIX had 1.7% fees and who knows how well they implemented their financing.
If the assertion is that a 3X fund should return 3X the index in the long term because volatility has no effect, then I would suggest you to learn about it a little more. Just sorry, but no.

This has nothing to do with the market being irrational. It simply has to do with how returns correspond to price movement when leverage is reset daily.

I rather contest the assertion that the volatility effect is not a fixed permanent condition. Simply stated, one can not short UPRO and go long SPY in proportion and be guaranteed anything.

For example, shorting UPRO and with 25% volatility when the index shows a 20% loss will lower your losses compared to 3X SPY. Similarly, shorting shorting UPRO and with 25% volatility when the index shows a 40% gain will increase your profits compared to 3X SPY.

In fact, I'd say that shorting UPRO and going long 3X spy helps you the most when volatility and returns are both high and hurts you the most when volatility is low and returns are either very good or very bad. In any case, with volatility often running between 10-25%, whether shorting UPRO and going long 3x SPY helps you or hurts you really depends on the returns and the sequence of returns and volatility on a daily basis.

If you do the math, you will see what I said is correct. The only way for 3X leveraged funds to return 3X the index is by adding money when it does poorly and removing when it does well. You can do this on a daily basis but you have to contend with a cash drag as you keep a pile of cash available. HFEA does it quarterly using 3X LTT.

For the strategy, if UPRO and TMF both need cash infusions the same time (i.e. stagflation), you might think about adding extra cash so you don't suffer a permanent loss.

I am really sorry, but I dismiss your contention and being factually incorrect as shown in 3X ETFs prospectus.

Image
I've done extensive research on the subject and I think you are not properly thinking through how shorting UPRO and buying 3x SPY would be a guaranteed risk-free return if what you are saying is true.
I think you should do more.
skierincolorado wrote: Fri Sep 17, 2021 7:02 pm UPRO has returned 30x since inception while SPY has returned 5x. That's 6x better not 3x. No volatility decay there, quite the opposite. Yes the market is mostly up, but there are ups and downs every day, and some big drops in there along the way too.
My assertion has never been that the volatility effect goes in one direction. Clearly, good returns at lower volatility will boost performance by using a 3X fund. That is what the prospectus asserts.
skierincolorado wrote: Fri Sep 17, 2021 7:02 pm SSO has returned nearly 2x SPY since inception in 2007. That's despite fees, borrowing costs, and a 50% market crash. On a pre-fee pre-borrowing cost measure, it's likely over 2x SPY - no volatility decay here either.

The classic example is ULPIX, but ULPIX had 1.7% fees, borrowing costs were higher in the 1990s, and we don't know how efficient its execution was.

You're simply incorrect. If volatility decay were a real theoretical construct, then it could be aribtrated and eliminated by hedge funds.
I am simply not incorrect. Period and end of story. Oh shall we look at what the prospectus says...
Compounding Risk — The Fund has a single day investment objective, and the Fund’s performance for any other period is the result of its return for each day compounded over the period. The performance of the Fund for periods longer than a single day will very likely differ in amount, and possibly even direction, from three times (3x) the daily return of the Index for the same period, before accounting for fees and expenses. Compounding affects all investments, but has a more significant impact on a leveraged fund. This effect becomes more pronounced as Index volatility and holding periods increase
Not sure what your "research" consists of, but portfolio visualizer and the Simba's backtesting sheet are not sufficient.

To maintain their investment objectives, geared funds rebalance their exposure to their underlying benchmarks each day by trimming or adding to their positions. As a result of daily fund rebalancing, an investor holding a geared fund longer-term is unlikely to continue to receive the fund's multiple times the benchmark's returns.

See EFFECTS OF DAILY REBALANCING AND COMPOUNDING ON GEARED INVESTING in the prospectus https://www.proshares.com/funds/prospec ... icker=UPRO
As long as the fund is held, compounding can cause the investor’s exposure to the underlying benchmark to continue to deviate from the fund's stated objective. In trending periods, compounding can enhance returns, but in volatile periods, compounding may hurt returns. Generally speaking, the greater the multiple or more volatile a fund's benchmark, the more pronounced the effects can be.
See SPECIAL NOTE REGARDING THE CORRELATION RISKS OF GEARED FUNDS in the SIA https://www.proshares.com/media/prospec ... 1923975093
As a result of compounding, for periods greater than one day, the use of leverage tends to cause the performance of a Fund to vary from its benchmark performance times the stated multiple or inverse multiple in the Fund’s investment objective, before accounting for fees and expenses. Compounding affects all investments, but has a more significant impact on the Geared Funds. Four factors significantly affect how close daily compounded returns are to longer-term benchmark returns times the fund’s multiple: the length of the holding period, benchmark volatility, whether the multiple is positive or inverse, and its leverage level. Longer holding periods, higher benchmark volatility, inverse exposure and greater leverage each can lead to returns that differ in amount, and possibly even direction, from a Geared Fund’s stated multiple times its benchmark return. As the tables below show, particularly during periods of higher benchmark volatility, compounding will cause longer term results to vary from the benchmark performance times the stated multiple in the Fund’s investment objective.
Feel free to believe what you want. I don't really care. When you post false information though, I will have to correct it. Sorry again if the ruffles the view of how you want things to be. Not my fault.
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: Why are leveraged funds (2x/3x) so expensive?

Post by skierincolorado »

typical.investor wrote: Fri Sep 17, 2021 7:19 pm
skierincolorado wrote: Fri Sep 17, 2021 7:02 pm
typical.investor wrote: Fri Sep 17, 2021 6:54 pm
skierincolorado wrote: Fri Sep 17, 2021 6:04 pm
typical.investor wrote: Fri Sep 17, 2021 5:56 pm
Because people need to know that the only way to offset volatility drag is by taking money out when it does well and putting more in when it doesn’t? That’s the only way you can earn 3X (minus costs).
Not really true. In theory, volatility drag shouldn't exist in the long-run. If volatility drag was a real, permanent feature, you could make guaranteed money by shorting UPRO and buying SPY in proportion. But it can exist in volatile markets and the market can remain irrational longer than you can wait, so I agree it's best to try to avoid.

SSO is getting pretty close to 2x SPY's return since 2007. If the market keeps going up with low volatility, SSO will surpass 2x SPY returns over the course of its life. And that's after fees and borrowing costs! It's probably already surpassed 2x SPY returns pre-fees pre-borrowing costs.

People like to reference ULPIX since the late 1990s - but ULPIX had 1.7% fees and who knows how well they implemented their financing.
If the assertion is that a 3X fund should return 3X the index in the long term because volatility has no effect, then I would suggest you to learn about it a little more. Just sorry, but no.

This has nothing to do with the market being irrational. It simply has to do with how returns correspond to price movement when leverage is reset daily.

I rather contest the assertion that the volatility effect is not a fixed permanent condition. Simply stated, one can not short UPRO and go long SPY in proportion and be guaranteed anything.

For example, shorting UPRO and with 25% volatility when the index shows a 20% loss will lower your losses compared to 3X SPY. Similarly, shorting shorting UPRO and with 25% volatility when the index shows a 40% gain will increase your profits compared to 3X SPY.

In fact, I'd say that shorting UPRO and going long 3X spy helps you the most when volatility and returns are both high and hurts you the most when volatility is low and returns are either very good or very bad. In any case, with volatility often running between 10-25%, whether shorting UPRO and going long 3x SPY helps you or hurts you really depends on the returns and the sequence of returns and volatility on a daily basis.

If you do the math, you will see what I said is correct. The only way for 3X leveraged funds to return 3X the index is by adding money when it does poorly and removing when it does well. You can do this on a daily basis but you have to contend with a cash drag as you keep a pile of cash available. HFEA does it quarterly using 3X LTT.

For the strategy, if UPRO and TMF both need cash infusions the same time (i.e. stagflation), you might think about adding extra cash so you don't suffer a permanent loss.

I am really sorry, but I dismiss your contention and being factually incorrect as shown in 3X ETFs prospectus.

Image
I've done extensive research on the subject and I think you are not properly thinking through how shorting UPRO and buying 3x SPY would be a guaranteed risk-free return if what you are saying is true.
I think you should do more.
skierincolorado wrote: Fri Sep 17, 2021 7:02 pm UPRO has returned 30x since inception while SPY has returned 5x. That's 6x better not 3x. No volatility decay there, quite the opposite. Yes the market is mostly up, but there are ups and downs every day, and some big drops in there along the way too.
My assertion has never been that the volatility effect goes in one direction. Clearly, good returns at lower volatility will boost performance by using a 3X fund. That is what the prospectus asserts.
skierincolorado wrote: Fri Sep 17, 2021 7:02 pm SSO has returned nearly 2x SPY since inception in 2007. That's despite fees, borrowing costs, and a 50% market crash. On a pre-fee pre-borrowing cost measure, it's likely over 2x SPY - no volatility decay here either.

The classic example is ULPIX, but ULPIX had 1.7% fees, borrowing costs were higher in the 1990s, and we don't know how efficient its execution was.

You're simply incorrect. If volatility decay were a real theoretical construct, then it could be aribtrated and eliminated by hedge funds.
I am simply not incorrect. Period and end of story. Oh shall we look at what the prospectus says...
Compounding Risk — The Fund has a single day investment objective, and the Fund’s performance for any other period is the result of its return for each day compounded over the period. The performance of the Fund for periods longer than a single day will very likely differ in amount, and possibly even direction, from three times (3x) the daily return of the Index for the same period, before accounting for fees and expenses. Compounding affects all investments, but has a more significant impact on a leveraged fund. This effect becomes more pronounced as Index volatility and holding periods increase
Not sure what your "research" consists of, but portfolio visualizer and the Simba's backtesting sheet are not sufficient.

To maintain their investment objectives, geared funds rebalance their exposure to their underlying benchmarks each day by trimming or adding to their positions. As a result of daily fund rebalancing, an investor holding a geared fund longer-term is unlikely to continue to receive the fund's multiple times the benchmark's returns.

See EFFECTS OF DAILY REBALANCING AND COMPOUNDING ON GEARED INVESTING in the prospectus https://www.proshares.com/funds/prospec ... icker=UPRO
As long as the fund is held, compounding can cause the investor’s exposure to the underlying benchmark to continue to deviate from the fund's stated objective. In trending periods, compounding can enhance returns, but in volatile periods, compounding may hurt returns. Generally speaking, the greater the multiple or more volatile a fund's benchmark, the more pronounced the effects can be.
See SPECIAL NOTE REGARDING THE CORRELATION RISKS OF GEARED FUNDS in the SIA https://www.proshares.com/media/prospec ... 1923975093
As a result of compounding, for periods greater than one day, the use of leverage tends to cause the performance of a Fund to vary from its benchmark performance times the stated multiple or inverse multiple in the Fund’s investment objective, before accounting for fees and expenses. Compounding affects all investments, but has a more significant impact on the Geared Funds. Four factors significantly affect how close daily compounded returns are to longer-term benchmark returns times the fund’s multiple: the length of the holding period, benchmark volatility, whether the multiple is positive or inverse, and its leverage level. Longer holding periods, higher benchmark volatility, inverse exposure and greater leverage each can lead to returns that differ in amount, and possibly even direction, from a Geared Fund’s stated multiple times its benchmark return. As the tables below show, particularly during periods of higher benchmark volatility, compounding will cause longer term results to vary from the benchmark performance times the stated multiple in the Fund’s investment objective.
Feel free to believe what you want. I don't really care. When you post false information though, I will have to correct it. Sorry again if the ruffles the view of how you want things to be. Not my fault.
The prospectus simply states that it varies. That is true. Frequently, as in the case of upro, it varies to the high side. However over very long periods theoretically there should be no deviation other than what is due to fees and borrowing costs. I implore you to think through why this must be the case given that it would be possible for hedge funds to exploit what you are saying for risk free returns until the phenomenon was eliminated. Your hypothesis amounts to market timing and a complete and total rejection of the efficient market hypothesis.

Furthermore, the actual returns of leveraged funds prove this to be the case. Sso has returned over 2x its benchmark after subtracting borrowing costs and fees over a 15 year period including multiple major market crashes.

Furthermore I have no bone in this fight as I do not and have not ever owned letf. I am simply providing the result of my research into forms of leverage. I personally decided to use other forms of leverage for other reasons (primarily fees).

Let me ask, why do you think sso has shown no volatility decay over 14 years of existence through all of the daily gyrations and two major market crashes? In fact eliminating fees alone gives sso 2x the return of spy. The cagr since 2007 is 14.21. If we add back .8 for the expense ratio which is .8 higher than spy, the cagr is 15.01. Compounded over 14.75 years that gives a return nearly exactly 2x spy. And we haven’t even considered borrowing costs yet which very conservatively would add another 1% to cagr. And again, this is through two crashes and all the other gyrations. And the cagr of spy over the period is very nears the 100+ year cagr (10%), so it’s not particularly great period for the market and has lots of volatility. So why no volatility decay since 2007 despite starting in a market peak? Actually the reverse of volatility decay.

The sharpe ratio of spy since 2007 is .67, near the modern average and again indicating this wasn’t some great period for the market. By comparison, the sharpe ratio from 2010-present is 1.05 - now that’s a great period for the market.

Do you can theorize it’s there. You can cite the prospectus, which I’ve already read. But it simply has not happened over a 15 year period of fairly typical market conditions.
Last edited by skierincolorado on Fri Sep 17, 2021 7:53 pm, edited 1 time in total.
User avatar
typical.investor
Posts: 5245
Joined: Mon Jun 11, 2018 3:17 am

Re: Why are leveraged funds (2x/3x) so expensive?

Post by typical.investor »

skierincolorado wrote: Fri Sep 17, 2021 7:25 pm However over very long periods theoretically there should be no deviation other than what is due to fees and borrowing costs. I implore you to think through why this must be the case give that it would be possible for hedge funds to exploit what you are saying for risk free returns.
Yeah and I implore you to think through it again.

Theoretically, why are volatility and returns certain to be in any particular sequence? They simply aren't.

As such, actual returns of a 3X leveraged funds are likely to differ from 3X the index.

The only way to assure that a 3X leveraged fund tracks 3X the index, is to remove $$ when the 3X fund does better and remove $$ when it does worse.

Let's have another look at the SAI graph as it has a little more detail.
Image

Please explain to me your theory.

In a flattish market (-0.5% to 5% returns) with sort of typical volatility (10-15%), how do hedge funds ensure that 3X funds won't have worse return that 3X the index?

How do you know in any holding period that we won't fall into the darker area (which is worse performance than 3X the index)?

In a completely sideways market (zero percent return), what is stopping you from realizing a loss?

There is simply no basis for the expectation that long term a 3X leveraged fund will return 3X the market. Returns are dependent on the sequence of volatility and index performance on a daily basis. Hedge funds are not operating to ensure a 3X leveraged fund will return 3X the market. The market has no underlying condition ensuring anything.

I really don't think one should be in fantasyland while making investments. The fact is, the SEC restricts 3X funds because they know and 3X fund may or may not return 3X the index and they want to protect investors who think they will because there is a 3X in the name.

If you have a theory of why the SEC is wrong, please contact them. I am sorry and can't imaging what you are thinking. The assertion that a 3X leveraged fund is sure to return 3X the index in the long run is simply wrong.
Last edited by typical.investor on Fri Sep 17, 2021 8:00 pm, edited 2 times in total.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Why are leveraged funds (2x/3x) so expensive?

Post by willthrill81 »

Higher costs = higher ERs
The Sensible Steward
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: Why are leveraged funds (2x/3x) so expensive?

Post by skierincolorado »

typical.investor wrote: Fri Sep 17, 2021 7:52 pm
skierincolorado wrote: Fri Sep 17, 2021 7:25 pm However over very long periods theoretically there should be no deviation other than what is due to fees and borrowing costs. I implore you to think through why this must be the case give that it would be possible for hedge funds to exploit what you are saying for risk free returns.
Yeah and I implore you to think through it again.

Theoretically, why are volatility and returns certain to be in any particular sequence? They simply aren't.

As such, actual returns of a 3X leveraged funds are likely to differ from 3X the index.

The only way to assure that a 3X leveraged fund tracks 3X the index, is to remove $$ when the 3X fund does better and remove $$ when it does worse.

Let's have another look at the SAI graph as it has a little more detail.
Image

Please explain to me your theory.

In a flattish market (-0.5% to 5% returns) with sort of typical volatility (10-15%), how do hedge funds ensure that 3X funds won't have worse return that 3X the index?

How do you know in any holding period that we won't fall into the darker area (which is worse performance than 3X the index)?

In a completely sideways market (zero percent return), what is stopping you from realizing a loss?

There is simply no basis for the expectation that long term a 3X leveraged fund will return 3X the market. Returns are dependent on the sequence of volatility and index performance on a daily basis. Hedge funds are not operating to ensure a 3X leveraged fund will return 3X the market.
In a flattish market with high volatility you are absolutely correct. There would be very significant volatility decay. The converse is also true though, in strong markets the returns will be much higher than the gearing ratio (and have been in actual funds). In reality, over very long periods these two phenomenon balance out both in theory and in the actual returns of funds.

The problem occurs when you assume over very long periods, across very long term market conditions, that volatility decay will occur. If this were the case then hedge funds could arbitrage this phenomenon out of existence.

I was editing my post a lot above to show how this has been the actual case for sso. It’s returns pre financing and pre expenses are significantly over 2x spy. This is in a fairly long period, starting and ending in market highs, with fairly typical market returns and volatility. The sharpe ratio of spy since 2007 is .67 which is only slightly above the modern average, and likely explains why sso returned more than 2x its benchmark.


There are good reasons for the sec to restrict 3x funds other than volatility decay. The investments are extremely risky. The fees and borrowing costs are high and not transparent. Volatility decay can occur over the short to intermediate term. Again, the actual returns of 2x and 3x funds are what you would expect if volatility decay did not exist.
User avatar
typical.investor
Posts: 5245
Joined: Mon Jun 11, 2018 3:17 am

Re: Why are leveraged funds (2x/3x) so expensive?

Post by typical.investor »

skierincolorado wrote: Fri Sep 17, 2021 8:00 pm
In a flattish market with high volatility you are absolutely correct. There would be very significant volatility decay. The converse is also true though, in strong markets the returns will be much higher than the gearing ratio (and have been in actual funds). In reality, over very long periods these two phenomenon balance out both in theory and in the actual returns of funds.
Not sure what theory states volatility and returns over the long term will occur in a sequence such that the effect will wash out. Do you have any source?
skierincolorado wrote: Fri Sep 17, 2021 8:00 pm The problem occurs when you assume over very long periods, across very long term market conditions, that volatility decay will occur. If this were the case then hedge funds could arbitrage this phenomenon out of existence.
I have made no such assumption.

My assertion is that 3X leveraged fund will return something different that 3X the index. Drag or boost? I don't and can't know and neither can hedge funds because we don't know future volatility and we don't know future returns and we don't know the sequence.

What we do know is that in a typical investor's holding period, that the returns of the two can be quite different. And we do know that the way to offset the volatility effect is to add when the 3X fund does worse than 3X the index and remove when the 3X fund does better. Daily, quarterly, yearly whatever. You can keep on track to 3X the underlying index by doing so.

The SEC has surely taken their stance in the belief that investors will assume a 3X fund will return 3X the index for their holding period. That is simply true for only one day, and if you don't adjust it from day 2, you may start to drift from 3X the index. Simply a fact.
User avatar
nisiprius
Advisory Board
Posts: 52105
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Why are leveraged funds (2x/3x) so expensive?

Post by nisiprius »

Expect more leveraged and inverse ETFs
On Wednesday, the Securities and Exchange Commission (SEC) passed a controversial new rule that, among other things, would make it significantly easier to launch leveraged and inverse exchange-traded funds, but at smaller multipliers than previously approved.
New 3X ETFs are blocked but 2X ETFs are eased. So perhaps we will see more competition among 2X ETFs?
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: Why are leveraged funds (2x/3x) so expensive?

Post by skierincolorado »

typical.investor wrote: Fri Sep 17, 2021 8:12 pm
skierincolorado wrote: Fri Sep 17, 2021 8:00 pm
In a flattish market with high volatility you are absolutely correct. There would be very significant volatility decay. The converse is also true though, in strong markets the returns will be much higher than the gearing ratio (and have been in actual funds). In reality, over very long periods these two phenomenon balance out both in theory and in the actual returns of funds.
Not sure what theory states volatility and returns over the long term will occur in a sequence such that the effect will wash out. Do you have any source?
skierincolorado wrote: Fri Sep 17, 2021 8:00 pm The problem occurs when you assume over very long periods, across very long term market conditions, that volatility decay will occur. If this were the case then hedge funds could arbitrage this phenomenon out of existence.
I have made no such assumption.

My assertion is that 3X leveraged fund will return something different that 3X the index. Drag or boost? I don't and can't know and neither can hedge funds because we don't know future volatility and we don't know future returns and we don't know the sequence.

What we do know is that in a typical investor's holding period, that the returns of the two can be quite different. And we do know that the way to offset the volatility effect is to add when the 3X fund does worse than 3X the index and remove when the 3X fund does better. Daily, quarterly, yearly whatever. You can keep on track to 3X the underlying index by doing so.

The SEC has surely taken their stance in the belief that investors will assume a 3X fund will return 3X the index for their holding period. That is simply true for only one day, and if you don't adjust it from day 2, you may start to drift from 3X the index. Simply a fact.
It’s the fact that if long term volatility decay were theoretically inherent in letf or daily rebalancing generally, hedge funds could arbitrage it away. Just think it through. Moreover long term returns of letf, minus fees and financing, empirically support this. You are correct that over medium term 5-10 years it may deviate significantly, up or down, and that this is a risk for investors with all but the longest time horizon
User avatar
nisiprius
Advisory Board
Posts: 52105
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Why are leveraged funds (2x/3x) so expensive?

Post by nisiprius »

If buying and holding a 3X leveraged ETF got you the same long-term results as getting 3X leverage with a margin loan--but without any risk of a margin call--that would be a free lunch.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: Why are leveraged funds (2x/3x) so expensive?

Post by skierincolorado »

nisiprius wrote: Fri Sep 17, 2021 9:03 pm If buying and holding a 3X leveraged ETF got you the same long-term results as getting 3X leverage with a margin loan--but without any risk of a margin call--that would be a free lunch.
No it’s not, for one there are large fees. Second, you can have a 3x leveraged position on s&p500 within a larger portfolio of bonds for example and not experience a margin call. For example a portfolio of 90k bonds 10k cash where I want to buy 30k of sp500. Do I buy upro or so I take 20k of margin? Neither method could result in a margin call. Third, investor using traditional broker margin near max margin are a tiny part of the market.

Fourth, and most importantly, actual results prove that this is what has happened. Before fees and financing, sso has returned over 2x spy, during a long 15 year period of typical market conditions. There is nothing magical about daily rebalancing that inherently lowers returns.
User avatar
Lee_WSP
Posts: 10346
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

Re: Why are leveraged funds (2x/3x) so expensive?

Post by Lee_WSP »

Portfolio visualizer has a new feature that lets you leverage The back test. You two can resolve this issue for at least the margin rates and time periods we have data available.
User avatar
typical.investor
Posts: 5245
Joined: Mon Jun 11, 2018 3:17 am

Re: Why are leveraged funds (2x/3x) so expensive?

Post by typical.investor »

skierincolorado wrote: Fri Sep 17, 2021 8:51 pm
typical.investor wrote: Fri Sep 17, 2021 8:12 pm
skierincolorado wrote: Fri Sep 17, 2021 8:00 pm
In a flattish market with high volatility you are absolutely correct. There would be very significant volatility decay. The converse is also true though, in strong markets the returns will be much higher than the gearing ratio (and have been in actual funds). In reality, over very long periods these two phenomenon balance out both in theory and in the actual returns of funds.
Not sure what theory states volatility and returns over the long term will occur in a sequence such that the effect will wash out. Do you have any source?
skierincolorado wrote: Fri Sep 17, 2021 8:00 pm The problem occurs when you assume over very long periods, across very long term market conditions, that volatility decay will occur. If this were the case then hedge funds could arbitrage this phenomenon out of existence.
I have made no such assumption.

My assertion is that 3X leveraged fund will return something different that 3X the index. Drag or boost? I don't and can't know and neither can hedge funds because we don't know future volatility and we don't know future returns and we don't know the sequence.

What we do know is that in a typical investor's holding period, that the returns of the two can be quite different. And we do know that the way to offset the volatility effect is to add when the 3X fund does worse than 3X the index and remove when the 3X fund does better. Daily, quarterly, yearly whatever. You can keep on track to 3X the underlying index by doing so.

The SEC has surely taken their stance in the belief that investors will assume a 3X fund will return 3X the index for their holding period. That is simply true for only one day, and if you don't adjust it from day 2, you may start to drift from 3X the index. Simply a fact.
It’s the fact that if long term volatility decay were theoretically inherent in letf or daily rebalancing generally, hedge funds could arbitrage it away. Just think it through. Moreover long term returns of letf, minus fees and financing, empirically support this. You are correct that over medium term 5-10 years it may deviate significantly, up or down, and that this is a risk for investors with all but the longest time horizon
I see you arbitrarily conjured up a palatable horizon there.

Again for the umpteenth time, nobody (in their right mind) is saying that decay is inherent. As such, hedge funds have no assurance that your arbitrage strategy will work.

All we do know is that the realized sequence of volatility and returns will mean a 3X leveraged fund will have different returns than 3X the index. Sure if hedge funds knew it would be a deficit, they could short it. They don't and can't.
skierincolorado wrote: Fri Sep 17, 2021 9:15 pm There is nothing magical about daily rebalancing that inherently lowers returns.
Again, nobody (who understand correctly) is saying there is.

We simply don't and can't know if a 3X leveraged fund will be over, under or the same as 3X market returns. So an investor using a 3X leveraged fund takes additional risk compared to a 3X market investor.

Assuming returns will be the same because *otherwise* hedge funds would arbitrage ignores the basic facts laid out in the prospectus that we don't know if returns will be better or worse. Hedge funds could kill themselves by trying to short UPRO and going 3X the market. Why are you assuming that they'd make out like bandits if a volatility effect exists? That same volatility effect which you say could make hedge funds rich, could also make them poor. Thus, no-one can arbitrage it away because it isn't known whether the effect will be a drag or boost.

If you using 3x leveraged funds (particularly UPRO and TMF in combination) my advice would be to add money if they simultaneously are doing worse than 3X their index. Relying on an assumption that a 3X leveraged fund will equate 3X the market return is simply not thinking things through.

And excuse me but saying that 3X leveraged funds must equate 3X the market return because hedge funds would arbitrage if they were sure of a volatility drag really misses the picture. Look again at the post charts. Do you see two colors. I apologize if you are color blind and maybe don't. Yeah, there are really two cases there. A theory that only accounts for one isn't very useful.
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: Why are leveraged funds (2x/3x) so expensive?

Post by skierincolorado »

typical.investor wrote: Fri Sep 17, 2021 10:00 pm

Again, nobody (who understand correctly) is saying there is.

We simply don't and can't know if a 3X leveraged fund will be over, under or the same as 3X market returns. So an investor using a 3X leveraged fund takes additional risk compared to a 3X market investor.

Assuming returns will be the same because *otherwise* hedge funds would arbitrage ignores the basic facts laid out in the prospectus that we don't know if returns will be better or worse. Hedge funds could kill themselves by trying to short UPRO and going 3X the market. Why are you assuming that they'd make out like bandits if a volatility effect exists? That same volatility effect which you say could make hedge funds rich, could also make them poor. Thus, no-one can arbitrage it away because it isn't known whether the effect will be a drag or boost.

If you using 3x leveraged funds (particularly UPRO and TMF in combination) my advice would be to add money if they simultaneously are doing worse than 3X their index. Relying on an assumption that a 3X leveraged fund will equate 3X the market return is simply not thinking things through.

And excuse me but saying that 3X leveraged funds must equate 3X the market return because hedge funds would arbitrage if they were sure of a volatility drag really misses the picture. Look again at the post charts. Do you see two colors. I apologize if you are color blind and maybe don't. Yeah, there are really two cases there. A theory that only accounts for one isn't very useful.
FIrst of all, your original assertion was that the only way to earn 3x was to put money in when the fund does well and take it out when it does poorly. This is simply not true as a matter of observable fact. The returns of SSO are very near 2x pre-fees and borrowing over a long 15 year period of varied market conditions.

Second, just because we don't know whether returns will be better or worse than the gearing ratio, doesn't mean that we and other market participants don't or shouldn't have an expectation of what it will be. If over an extended period, returns of daily rebalancing (or LETF using daily rebalancing) were significantly better (or worse), market participants would go long (or short) on the strategy. Thus we can expect over extended periods that returns of daily rebalancing, or LETF that use daily rebalancing, will have returns near their gearing ratio.

Let's say, in simulations, a hedge fund finds that over the last 10 years daily rebalancing has consistently underperformed the gearing ratio - not just by a little bit, but by some statistically significant measure. They would short the strategy of daily rebalancing and the phenomenon would disappear. Likewise, if they found over the last 10 years there was a highly statistically significant overperformance of daily rebalancing - they would go long the strategy and start daily rebalancing themselves. Thus over long enough periods, we can expect that hedge funds and other market participants are analyzing these trends and acting on them in ways that ensure the long-term returns of daily rebalancing are reasonably close to the gearing ratio - minus fees and expenses.

If all you are saying is that over shorter periods daily rebalancing may deviate from gearing ratio, well I've already said that multiple times already, yes it is an additional risk over short and medium time horizons, and notwithstanding the aspersions of color blindness, we are in agreement. That's no different than saying that the market may go up 25% or down 25% but we expect it will go up 8% on average. Just because we don't know what the market will do doesn't mean we don't have an expectation of future returns. Daily rebalancing may do better or worse than the gearing ratio, but we expect it will do the same and this expectation becomes stronger with less uncertainty the longer the time horizon. Any other expectation is market timing and a violation of the efficient market hypothesis.

(Bolding the best explanation I can come up with)
User avatar
typical.investor
Posts: 5245
Joined: Mon Jun 11, 2018 3:17 am

Re: Why are leveraged funds (2x/3x) so expensive?

Post by typical.investor »

skierincolorado wrote: Fri Sep 17, 2021 10:28 pm
typical.investor wrote: Fri Sep 17, 2021 10:00 pm

Again, nobody (who understand correctly) is saying there is.

We simply don't and can't know if a 3X leveraged fund will be over, under or the same as 3X market returns. So an investor using a 3X leveraged fund takes additional risk compared to a 3X market investor.

Assuming returns will be the same because *otherwise* hedge funds would arbitrage ignores the basic facts laid out in the prospectus that we don't know if returns will be better or worse. Hedge funds could kill themselves by trying to short UPRO and going 3X the market. Why are you assuming that they'd make out like bandits if a volatility effect exists? That same volatility effect which you say could make hedge funds rich, could also make them poor. Thus, no-one can arbitrage it away because it isn't known whether the effect will be a drag or boost.

If you using 3x leveraged funds (particularly UPRO and TMF in combination) my advice would be to add money if they simultaneously are doing worse than 3X their index. Relying on an assumption that a 3X leveraged fund will equate 3X the market return is simply not thinking things through.

And excuse me but saying that 3X leveraged funds must equate 3X the market return because hedge funds would arbitrage if they were sure of a volatility drag really misses the picture. Look again at the post charts. Do you see two colors. I apologize if you are color blind and maybe don't. Yeah, there are really two cases there. A theory that only accounts for one isn't very useful.
FIrst of all, your original assertion was that the only way to earn 3x was to put money in when the fund does well and take it out when it does poorly. This is simply not true theoretically or empirically. The returns of SSO are very near 2x pre-fees and borrowing over a long 15 year period of varied market conditions.
Well, I never meant to say that it couldn't happen. To ensure it does though ... yeah you gotta rebalance.

OK, theoretically and empirically then ... OK fine.

The question was address in research trying to determine if 3X leveraged funds increased volatility.

First, it states:
We show theoretically, however, that capital flows can lower ETF rebalancing demand and completely eliminate it in the limit
This simply means that if investors, as I said, adds $$ when the 3X leveraged fund underperforms and removes $$ when it outperforms that the fund itself would have no need to rebalance.

Next, it states:
Empirically, we find that capital flows substantially reduce ETF rebalancing demand, even during periods of severe market stress.

What that means is investors already are adding and removing based on market movements.

Read the paper but here are the basics:
To clearly illustrate the mechanism, we introduce a representative investor who wishes to track m-times the return on the index over a horizon longer than one period and assume that trading by the investor gives rise to capital flows.
First consider the case in which the investor does not rebalance his own portfolio to account for changes in the ETF’s exposure to the index.
holding the ETF for more than one period without rebalancing his portfolio generally will not provide the investor with a return that equals m-times the performance of the index over the longer holding period. This mismatch of returns is due to the fact that the return process for the ETF differs from the index return process.
the investor generally must rebalance his portfolio every period to successfully track m-times the return on the index over a longer time horizon.
Some possible explanations as to why capital flows do not entirely offset rebalancing demand could include market frictions that impede capital flows or the use of leveraged and inverse ETFs by some investors to implement momentum trading strategies.
Capturing momentum isn't actually that easy I think. And trying to short it when other are erroneously pursuing it isn't either. Otherwise it would be arbitraged away.
skierincolorado wrote: Fri Sep 17, 2021 10:28 pm Second, just because we don't know whether returns will be better or worse than the gearing ratio, doesn't mean that we and other market participants don't or shouldn't have an expectation of what it will be.
OK, but does that mean they can know well enough to take advantage of that?
skierincolorado wrote: Fri Sep 17, 2021 10:28 pm If over an extended period, returns of daily rebalancing (or LETF using daily rebalancing) were significantly better (or worse), market participants would go long (or short) on the strategy. Thus we can expect over extended periods that returns of daily rebalancing, or LETF that use daily rebalancing, will have returns near their gearing ratio.
Sure, and if over any extended period a momentum strategy were significantly better (or worse), market participants would go long (or short) on the strategy. I guess your argument then rests on the use of these funds as momentum strategies to neither be able to succeed nor fail.
skierincolorado wrote: Fri Sep 17, 2021 10:28 pm Let's say, in simulations, a hedge fund finds that over the last 10 years daily rebalancing has consistently underperformed the gearing ratio - not just by a little bit, but by some statistically significant measure. They would short the strategy of daily rebalancing and the phenomenon would disappear. Likewise, if they found over the last 10 years there was a highly statistically significant overperformance of daily rebalancing - they would go long the strategy and start daily rebalancing themselves. Thus over long enough periods, we can expect that hedge funds and other market participants are analyzing these trends and acting on them in ways that ensure the long-term returns of daily rebalancing are reasonably close to the gearing ratio - minus fees and expenses.
Sure, and in theory you'd expect momentum to be arbitraged away too.
skierincolorado wrote: Fri Sep 17, 2021 10:28 pm If all you are saying is that over shorter periods daily rebalancing may deviate from gearing ratio, well I've already said that multiple times already, yes it is an additional risk over short and medium time horizons, and notwithstanding the aspersions of color blindness, we are in agreement. That's no different than saying that the market may go up 25% or down 25% but we expect it will go up 8% on average. Daily rebalancing may do better or worse than the gearing ratio, but we expect it will do the same and this expectation becomes stronger with less uncertainty the longer the time horizon. Any other expectation is market timing and a violation of the efficient market hypothesis.
No, what I am saying is that theoretically and empirically that unless you employ capital flows to counter the direction of 3X leveraged fund outperformance (underperformance) that you will effectively be adopting a momentum strategy that may or may not work in your favor. Such a strategy typically struggles in a dramatic change of direction.

So yes, I will agree with you that if change of direction is not a risk, then surely momentum strategies will be arbitraged away. But I don't think the direction of gearing returns is going to be so constant. And I don't think that long term what effectively is a momentum strategy is going to be necessarily in line with a multiple of market returns.

What's to stop hedge funds from jumping into excess geared returns just before they turn poor and jumping out just before they turn good again. Of course if they could do the opposite, they could arbitrage. But again, it's a two way street and momentum is fickle.

https://papers.ssrn.com/sol3/papers.cfm ... id=2504012
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: Why are leveraged funds (2x/3x) so expensive?

Post by skierincolorado »

typical.investor wrote: Fri Sep 17, 2021 11:37 pm
skierincolorado wrote: Fri Sep 17, 2021 10:28 pm
typical.investor wrote: Fri Sep 17, 2021 10:00 pm

Again, nobody (who understand correctly) is saying there is.

We simply don't and can't know if a 3X leveraged fund will be over, under or the same as 3X market returns. So an investor using a 3X leveraged fund takes additional risk compared to a 3X market investor.

Assuming returns will be the same because *otherwise* hedge funds would arbitrage ignores the basic facts laid out in the prospectus that we don't know if returns will be better or worse. Hedge funds could kill themselves by trying to short UPRO and going 3X the market. Why are you assuming that they'd make out like bandits if a volatility effect exists? That same volatility effect which you say could make hedge funds rich, could also make them poor. Thus, no-one can arbitrage it away because it isn't known whether the effect will be a drag or boost.

If you using 3x leveraged funds (particularly UPRO and TMF in combination) my advice would be to add money if they simultaneously are doing worse than 3X their index. Relying on an assumption that a 3X leveraged fund will equate 3X the market return is simply not thinking things through.

And excuse me but saying that 3X leveraged funds must equate 3X the market return because hedge funds would arbitrage if they were sure of a volatility drag really misses the picture. Look again at the post charts. Do you see two colors. I apologize if you are color blind and maybe don't. Yeah, there are really two cases there. A theory that only accounts for one isn't very useful.
FIrst of all, your original assertion was that the only way to earn 3x was to put money in when the fund does well and take it out when it does poorly. This is simply not true theoretically or empirically. The returns of SSO are very near 2x pre-fees and borrowing over a long 15 year period of varied market conditions.
Well, I never meant to say that it couldn't happen. To ensure it does though ... yeah you gotta rebalance.

OK, theoretically and empirically then ... OK fine.

The question was address in research trying to determine if 3X leveraged funds increased volatility.

First, it states:
We show theoretically, however, that capital flows can lower ETF rebalancing demand and completely eliminate it in the limit
This simply means that if investors, as I said, adds $$ when the 3X leveraged fund underperforms and removes $$ when it outperforms that the fund itself would have no need to rebalance.

Next, it states:
Empirically, we find that capital flows substantially reduce ETF rebalancing demand, even during periods of severe market stress.

What that means is investors already are adding and removing based on market movements.

Read the paper but here are the basics:
To clearly illustrate the mechanism, we introduce a representative investor who wishes to track m-times the return on the index over a horizon longer than one period and assume that trading by the investor gives rise to capital flows.
First consider the case in which the investor does not rebalance his own portfolio to account for changes in the ETF’s exposure to the index.
holding the ETF for more than one period without rebalancing his portfolio generally will not provide the investor with a return that equals m-times the performance of the index over the longer holding period. This mismatch of returns is due to the fact that the return process for the ETF differs from the index return process.
the investor generally must rebalance his portfolio every period to successfully track m-times the return on the index over a longer time horizon.
Some possible explanations as to why capital flows do not entirely offset rebalancing demand could include market frictions that impede capital flows or the use of leveraged and inverse ETFs by some investors to implement momentum trading strategies.
Capturing momentum isn't actually that easy I think. And trying to short it when other are erroneously pursuing it isn't either. Otherwise it would be arbitraged away.
skierincolorado wrote: Fri Sep 17, 2021 10:28 pm Second, just because we don't know whether returns will be better or worse than the gearing ratio, doesn't mean that we and other market participants don't or shouldn't have an expectation of what it will be.
OK, but does that mean they can know well enough to take advantage of that?
skierincolorado wrote: Fri Sep 17, 2021 10:28 pm If over an extended period, returns of daily rebalancing (or LETF using daily rebalancing) were significantly better (or worse), market participants would go long (or short) on the strategy. Thus we can expect over extended periods that returns of daily rebalancing, or LETF that use daily rebalancing, will have returns near their gearing ratio.
Sure, and if over any extended period a momentum strategy were significantly better (or worse), market participants would go long (or short) on the strategy. I guess your argument then rests on the use of these funds as momentum strategies to neither be able to succeed nor fail.
skierincolorado wrote: Fri Sep 17, 2021 10:28 pm Let's say, in simulations, a hedge fund finds that over the last 10 years daily rebalancing has consistently underperformed the gearing ratio - not just by a little bit, but by some statistically significant measure. They would short the strategy of daily rebalancing and the phenomenon would disappear. Likewise, if they found over the last 10 years there was a highly statistically significant overperformance of daily rebalancing - they would go long the strategy and start daily rebalancing themselves. Thus over long enough periods, we can expect that hedge funds and other market participants are analyzing these trends and acting on them in ways that ensure the long-term returns of daily rebalancing are reasonably close to the gearing ratio - minus fees and expenses.
Sure, and in theory you'd expect momentum to be arbitraged away too.
skierincolorado wrote: Fri Sep 17, 2021 10:28 pm If all you are saying is that over shorter periods daily rebalancing may deviate from gearing ratio, well I've already said that multiple times already, yes it is an additional risk over short and medium time horizons, and notwithstanding the aspersions of color blindness, we are in agreement. That's no different than saying that the market may go up 25% or down 25% but we expect it will go up 8% on average. Daily rebalancing may do better or worse than the gearing ratio, but we expect it will do the same and this expectation becomes stronger with less uncertainty the longer the time horizon. Any other expectation is market timing and a violation of the efficient market hypothesis.
No, what I am saying is that theoretically and empirically that unless you employ capital flows to counter the direction of 3X leveraged fund outperformance (underperformance) that you will effectively be adopting a momentum strategy that may or may not work in your favor. Such a strategy typically struggles in a dramatic change of direction.

So yes, I will agree with you that if change of direction is not a risk, then surely momentum strategies will be arbitraged away. But I don't think the direction of gearing returns is going to be so constant. And I don't think that long term what effectively is a momentum strategy is going to be necessarily in line with a multiple of market returns.

What's to stop hedge funds from jumping into excess geared returns just before they turn poor and jumping out just before they turn good again. Of course if they could do the opposite, they could arbitrage. But again, it's a two way street and momentum is fickle.

https://papers.ssrn.com/sol3/papers.cfm ... id=2504012
You've brought up the momentum factor to demonstrate that markets are not perfectly efficient. This is why I've always used words like "in an efficient market" and "reasonably close to the gearing ratio." And as I've shown even over long periods like 15 years for SSO during typical market conditions (not a great market) it has exeeded the gearing ratio slightly on a pre-tax pre-fee basis. I didn't say it perfectly matched it. Nevertheless it would be a remarkable coincidence for the returns to match the gearing ratio so closely if the relationship between returns and volatility were random. So my argument has never been that it would perfectly match it and has always been dependent on markets being efficient.


"What's to stop hedge funds from jumping into excess geared returns just before they turn poor and jumping out just before they turn good again. "

Such behavior is exactly what ensures there is little premium or cost to daily rebalancing. Hedge funds jumping into daily balancing strategies when they are good will increase daily volatility and eliminate the daily rebalancing bump. Jumping out when daily balancing turns bad will reduce daily market volatility and eliminate the daily rebalancing drag. If there was a long-term premium for daily rebalancing, more would do it. If there was a long-term cost to daily rebalancing, hedge funds could short daily rebalancing by doing the opposite (buying on all market dips and selling on all market ups within an overall hedged position). In a perfectly efficient market, as soon a statistical trends like this become highly significant they are eliminated.

Given the long-term return of LETF can be expected to be reasonably close to the gearing ratio in a reasonably efficient market, it is perfectly acceptable for the long-term investor to hold LETF without fighting the daily rebalance. There is some additional risk over short and medium time horizons that could be eliminated by fighting the rebalance.
Last edited by skierincolorado on Sat Sep 18, 2021 11:12 am, edited 1 time in total.
JackoC
Posts: 4710
Joined: Sun Aug 12, 2018 11:14 am

Re: Why are leveraged funds (2x/3x) so expensive?

Post by JackoC »

alex_686 wrote: Fri Sep 17, 2021 5:02 pm
nisiprius wrote: Fri Sep 17, 2021 4:28 pm Something resembling a loan needs to be negotiated every day. And the daily rebalancing seemingly means that stocks need to be bought or sold every day. Maybe those expenses add up.
They use futures. So no negotiation on margin loans or things like that. If they did use margin it would be counted as a expense. But not the implied cost of carry for futures.
A lot of them also use total return swaps* where the same point applies, or might. You have to first analyze what's being included as 'expense' which will sometimes create form over substance confusion. To further clarify your point, you can't directly compare the ER's of two funds one of which includes explicit interest cost in ER, one of which doesn't include futures implied borrowing cost in ER. The cost comes out of pre-expense return for the second fund (though might be a different amount).

Also someone mentioned transactions costs from rebalancing but that would generally not be included in ER, and difficult to find out directly. There's a lot going on, volatility drag also if trying to divine what expenses were via looking at past returns. Whereas if you stick to a broad index and DIY the futures to leverage (which might or might not suit your goals and risk tolerance, but here we're talking about LETF's and if leverage didn't suit you you wouldn't do either, end of discussion) you can relatively straight forwardly figure out the implied borrowing cost, your commission and bid-offer costs are directly visible, and rebalancing at your discretion. Helps a lot if you can do this in tax deferred space though. But in any case non-transparency is a big issue for LETF's, as IMO demonstrated all the more by various long threads about them (some arguing 'back tests show' volatility drag isn't real, yeah I don't think so :happy ): it's easy to go wrong looking for those answers.

*a dealer pays the fund 'total return of asset X', the fund pays 'short term benchmark interest rate plus spread', every short period like say 3 months for multiple such periods. This is common where the LETF index isn't a broad one like S&P, NASDAQ 100 etc where there are liquid futures contracts. The dealer might just hold and finance the assets as their hedge, or they might have a counterparty that wants to do the offsetting swap, like a -X LETF. A fund family with both X's and -X's in its product line up might do these offsetting swaps internally as part of positioning both types of fund, using external TRS, futures or financing/shorting the cash underlying for daily fine tuning of the exact amount.
User avatar
nisiprius
Advisory Board
Posts: 52105
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Why are leveraged funds (2x/3x) so expensive?

Post by nisiprius »

I obtained monthly returns for compensated for fund expenses by adding 1.60%/12 to each month's return for
ULPIX, the ProFunds UltraBull Fund,
SSO, the ProShares S&P 500 2X Ultra ETF.
These leveraged products, both from the same firm, both apply daily 2X leverage to the S&P 500 index.

skierincolorado has objected to the use of ULPIX as an illustration of leveraged fund behavior, because of high fund expenses. So I compensated for fund expenses by adding 1.60%/12 to each month's return for ULPIX and 0.91%/12 to each month's return for SSO, the ProShares S&P 500 2X Ultra ETF.

We see that when we do this, since inception of SSO, ULPIX+1.60% and SSO+0.91% tracked fairly closely together, showing that we have successfully compensated for the different expenses of the two products.

We also see that, as stated by skierincolorado, over the specific time period since inception of SSO, if they had had no expenses, either ULPIX or SSO would have delivered more than 2X the total cumulative return of the S&P 500. Very slightly more.

Image

We can plot this in another way. We can plot the multiple of the S&P 500 total return that expenseless ULPIX and SSO would have delivered, cumulatively, starting at inception of SSO in 2006. After some violent excursions, after about 2013 the multiple settles down. By 2014 it is finally beating the S&P 500, and the multiple gradually rises to the hoped-for 2X and in recent months finally exceeds it.

Image

Having shown that after we compensate for expenses, ULPIX and SSO track fairly closely together, it now becomes reasonable to look at ULPIX's history before 2006, knowing that we are not looking at expense effects, but only at sequence-of-return effects.

Image

Image

In this case, after compensating for expenses, ULPIX is behind the unleveraged 1X S&P 500 for over twenty years, and as of August 2021 would still have delivered only 1.4X the total cumulative return of the S&P 500, not even close to the hoped-for 2X.

Someone investing in either ULPIX or SSO, even if they could have done so with a zero expense ratio, from either of these inception dates, would have typically seen cumulative returns that were wildly different from 2X the S&P 500, and in the case of ULPIX, would still, today, far short of 2X.

And since we have removed expenses, we are simply looking at the effect of the sequences of returns, endpoints, and time ranges. The differences between SSO starting at inception of SSO, and ULPIX starting at inception of ULPIX, are mostly the result of different time frames, not differences in expenses.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
User avatar
Lee_WSP
Posts: 10346
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

Re: Why are leveraged funds (2x/3x) so expensive?

Post by Lee_WSP »

Isn't the issue with asking about 2x the returns always going to be because it's always a snap shot in time?

If you borrow 100% at the start date, I'm pretty certain you'll hardly ever match 2x the end date.
User avatar
typical.investor
Posts: 5245
Joined: Mon Jun 11, 2018 3:17 am

Re: Why are leveraged funds (2x/3x) so expensive?

Post by typical.investor »

nisiprius wrote: Sat Sep 18, 2021 5:41 pm Image

In this case, after compensating for expenses, ULPIX is behind the unleveraged 1X S&P 500 for over twenty years, and as of August 2021 would still have delivered only 1.4X the total cumulative return of the S&P 500, not even close to the hoped-for 2X.

Someone investing in either ULPIX or SSO, even if they could have done so with a zero expense ratio, from either of these inception dates, would have typically seen cumulative returns that were wildly different from 2X the S&P 500, and in the case of ULPIX, would still, today, far short of 2X.
Very nice work Nisi!!!!

More empirical evidence that as expected under theory, that the long term returns of a leveraged fund doesn't necessarily approximate a return of the corresponding multiple of the index.

To ensure an approximation, one must adjust the amount of exposure via cashflow (by removing exposure when the fund out performs and adding exposure when it under performs).
MIretired
Posts: 1059
Joined: Fri Sep 06, 2013 12:35 pm

Re: Why are leveraged funds (2x/3x) so expensive?

Post by MIretired »

Deleted.
User avatar
nisiprius
Advisory Board
Posts: 52105
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Why are leveraged funds (2x/3x) so expensive?

Post by nisiprius »

An important detail on my "multiple" chart, by the way, is that the multiplication factor applies to both gains and losses. On the ULPIX chart, for example... reducing the vertical scale... neither of the huge spikes was good news for ULPIX investors.

These are charts of overall cumulative gains or losses from the original $10,000 investment, comparing (adjusted costless) ULPIX to an unleveraged S&P 500 investment.

Image

The +50X spike in 2002 represents a time when an original $10,000 investment in the S&P 500 was down to $9,910, thus a $90 loss, while ULPIX was down to $5,483, thus a -$4,517 loss or 50X as big a loss.

The -50X spike in 2009 represents a time when the S&P 500 investment was up to $10,126, thus a $126 gain, while ULPIX had another staggering loss--down to $3,416 or a -$6,584 loss. Thus in this case, the cumulative total return of ULPIX was minus 50X the S&P 500.

Huge excursions in the multiple can be expected any time the growth chart for the S&P 500 is close to the $10,000 starting point, i.e. the S&P 500 has roughly broken even... because over periods of time much longer than a day, when the S&P 500 has cumulatively broken even, the leveraged fund has not been close breaking even.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
skierincolorado
Posts: 2377
Joined: Sat Mar 21, 2020 10:56 am

Re: Why are leveraged funds (2x/3x) so expensive?

Post by skierincolorado »

nisiprius wrote: Sat Sep 18, 2021 5:41 pm I obtained monthly returns for compensated for fund expenses by adding 1.60%/12 to each month's return for
ULPIX, the ProFunds UltraBull Fund,
SSO, the ProShares S&P 500 2X Ultra ETF.
These leveraged products, both from the same firm, both apply daily 2X leverage to the S&P 500 index.

skierincolorado has objected to the use of ULPIX as an illustration of leveraged fund behavior, because of high fund expenses. So I compensated for fund expenses by adding 1.60%/12 to each month's return for ULPIX and 0.91%/12 to each month's return for SSO, the ProShares S&P 500 2X Ultra ETF.

We see that when we do this, since inception of SSO, ULPIX+1.60% and SSO+0.91% tracked fairly closely together, showing that we have successfully compensated for the different expenses of the two products.

We also see that, as stated by skierincolorado, over the specific time period since inception of SSO, if they had had no expenses, either ULPIX or SSO would have delivered more than 2X the total cumulative return of the S&P 500. Very slightly more.

Image

We can plot this in another way. We can plot the multiple of the S&P 500 total return that expenseless ULPIX and SSO would have delivered, cumulatively, starting at inception of SSO in 2006. After some violent excursions, after about 2013 the multiple settles down. By 2014 it is finally beating the S&P 500, and the multiple gradually rises to the hoped-for 2X and in recent months finally exceeds it.

Image

Having shown that after we compensate for expenses, ULPIX and SSO track fairly closely together, it now becomes reasonable to look at ULPIX's history before 2006, knowing that we are not looking at expense effects, but only at sequence-of-return effects.

Image

Image

In this case, after compensating for expenses, ULPIX is behind the unleveraged 1X S&P 500 for over twenty years, and as of August 2021 would still have delivered only 1.4X the total cumulative return of the S&P 500, not even close to the hoped-for 2X.

Someone investing in either ULPIX or SSO, even if they could have done so with a zero expense ratio, from either of these inception dates, would have typically seen cumulative returns that were wildly different from 2X the S&P 500, and in the case of ULPIX, would still, today, far short of 2X.

And since we have removed expenses, we are simply looking at the effect of the sequences of returns, endpoints, and time ranges. The differences between SSO starting at inception of SSO, and ULPIX starting at inception of ULPIX, are mostly the result of different time frames, not differences in expenses.
This is very nice work, thank you. However there are two problems:

1) We still need to add back in borrowing cost. The borrowing cost is on only half the fund's assets. So if LIBOR is 4% in 2002, the borrowing cost for the fund is 2% (or 2/12 monthly). I'd suggest as a simplification breaking the fund's history into halves, or quarters, and adding the average borrowing cost for that period to the monthly returns. When I've done this the returns of ULPIX since 1998 get close to 2x. This has always been my contention that pre-fees AND pre-borrowing costs, 2x funds can be expected to return near their gearing ratio in an reasonably efficient market.

2) 1998 is a pretty terrible start date. It's quite a long period I agree, but it's also an exceptionally poor 23 year period for the market. The sharpe ratio of the S&P500 since 1998 is like 0.4 - well below the long-term average. I agree LETFs will do worse than their gearing ratio when returns are low and volatility high, and better than their gearing ratio when returns are high and volatility low. Since in the long run we can expect average returns and average volatility, we can also expect the funds to return near their gearing ratio.


After factoring in #1 you should get close to 2x for ULPIX anyways. I did in my calculations. Any remaining difference, if any, would be because this is a terrible start date. Lots of analyses blow up when a late 90s start date is picked because it's right before two crashes.
User avatar
nisiprius
Advisory Board
Posts: 52105
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Why are leveraged funds (2x/3x) so expensive?

Post by nisiprius »

skierincolorado wrote: Sun Sep 19, 2021 7:51 am...After factoring in #1 you should get close to 2x for ULPIX anyways. I did in my calculations. Any remaining difference, if any, would be because this is a terrible start date. Lots of analyses blow up when a late 90s start date is picked because it's right before two crashes...
I believe "since inception"--all available data, for real-world funds running real money with real expenses is the closest thing to unbiased one can be.

But my whole point is that ULPIX

1) most emphatically did not double the return of the S&P 500, and
2) it was not primarily due to fund expenses, although they probably were responsible for it literally underperforming the S&P 500.

As for "adding back in borrowing cost" I am darned if I can think of any justification for doing that.

Anyone who thinks they can double the return of the S&P 500 by buying and holding a 2X leveraged fund is likely to be severely disappointed. Their expectation will only be met if they can time the purchase for the start of a long, steady bull market.

Why not believe Direxion itself?
Q. Are Direxion ETFs Right for You?

A...Definitely not if you are a conservative investor who:
  • ...
  • Is unfamiliar with the unique nature and performance characteristics of funds which seek leveraged daily investment results
  • Is unable to manage your portfolio actively and make changes as market conditions and fund performance dictate.
    ...
"The unique nature and performance characteristics" is exactly what we are discussing, and they are unique, and are obviously hard to understand since they are as debated as the Monty Hall problem. Direxion says flatly that a leveraged ETF investor needs to manage their portfolio "actively." That means you have to believe you can time the market to make it work, you can't expect mechanical buying-and-holding to deliver that hoped-for 2X.

They also say:
Q. If the target index is up 10% for a month, shouldn't I expect to have a 30% gain in my Direxion Bull 3X ETF?

A. No, not typically.
and
Q. Are Direxion Shares ETFs appropriate for buy and hold investing?

A. No, this is not recommended. Leveraged ETFs seek daily investment results and should therefore be considered primarily for short-term trading purposes. [And special-case exceptions]
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
Post Reply