Beating the 7% Investment Goal with Leveraged ETFs

[specialkayme]

I've read (not just here, but just about everywhere) about how to achieve 7% average investment returns, after inflation, by investing in the stock market (SPY if you're a fan of the S&P 500, QQQ if you're a fan of the Dow, or VTI/VTSMX if you're a fan of the entire stock market). The key being that the market on the whole returns an average of 7% per year, after inflation, but each individual stock may or may not beat that average. So investing in broad based funds gives you the best chance of hitting that average (with low expense ratios to boot).

So why not use leveraged ETFs to accomplish a better return?

Obviously, leveraged ETFs are not for everyone. Granted. SPY, for example, isn't for everyone. Even the stock market itself isn't for everyone. But if you get past those issues, and we come to the conclusion that you will invest in the stock market, long term, and you will invest in broad based ETFs or mutual funds, why ignore the leveraged ETFs?

A few givens:
1. You save $X per paycheck.
2. You put that savings into a market account (IRA or Scottrade Account, or similar device).
3. You use that market account to purchase broad based ETFs or mutual funds.
4. You don't worry about the increase or decrease in the market, you just keep your head down and keep saving and investing, knowing over the long term you'll return a 7% average if you invest long enough.
5. Keeping that mentality, you determine SPY or QQQ is the best investment for you.

If that's the case, why not use UPRO, or UDOW, triple your earnings?

Yes, it may go down, but you've already decided that you'll stick it out and keep investing anyway. So why not live on the larger upside of a 3x leveraged ETF?

I hear everyone talk about buying SPY, QQQ, or VTI, but you don't hear about UPRO or UDOW. Other than larger risk, why not?

[secondcor521]

I hear everyone talk about buying SPY, QQQ, or VTI, but you don't hear about UPRO or UDOW. Other than larger risk, why not?

Last I looked, larger expense ratios as well.

I went down the same thought path you did, and for me, it seemed like once one got past 100% stocks, the expenses and risks of the leverage meant that the risk/reward ratio took a significant turn for the worse.

I also decided that I would rather accelerate FIRE through increasing my savings rate rather than trying to increase my investment returns.

But yeah, I was basically 100% stocks from 1987 through about 2015 and it worked out just as you describe.

Oh, and back to your step 4.  I watched my portfolio go down in 2000/2001 and also in 2008/2009, and even with my tolerance for risk it was somewhat of a mental challenge to stick with the game plan.  Someone who would hang on in and through a downturn with 100% stocks might not be able to stomach 200% or 300% stocks.

[specialkayme]

I went down the same thought path you did, and for me, it seemed like once one got past 100% stocks, the expenses and risks of the leverage meant that the risk/reward ratio took a significant turn for the worse.

I haven't quite gotten there . . . yet . . .

Can you explain to me how you came to the conclusion the risk/reward ratio takes a signifcant turn for the worse?

I also decided that I would rather accelerate FIRE through increasing my savings rate rather than trying to increase my investment returns.

Who said you can't do both?

Someone who would hang on in and through a downturn with 100% stocks might not be able to stomach 200% or 300% stocks.

But massive downturns in the market I think is where you're really going to get the advantage of a leveraged ETF. Granted, my mind may be moving faster than my math, which is what imposed this question, but here's what I'm thinking.

Lets say the S&P increases 30% over three years, equally each day, which would make for a 9.13% annualized gain per year (I wish I could calculate the daily return here, but I'm tired).
Leveraged ETFs do 2x or 3x (depending on the fund you get) each day's market, meaning in this instance UPRO would increase by 27.39% per trading year, for a 107% gain in total.

Now, lets say the S&P dropped 30% over three years. That would make for a 11.19% annualized loss on the S&P. But trippling that loss on UPRO would give you a 33.57% annualized loss, which would make for a 71% loss overall.

With gains, you're gaining more than 3x the S&P's gains, but losing less than 3x the S&P's losses. At least, that's what my math shows.

And assume the S&P went down 50% in one month. UPRO can't physically go down 150% in that same month. But if the S&P went up 50% in one month, UPRO sure could go up 150% that month. So it seems like you have significant downside protection, with upside potential. That's assuming you were going to invest in the S&P anyway.

So, where did my math go wrong? I know it must have somewhere, or everyone would be talking about leveraged ETFs. . . .

[kendallf]

The compounding actually penalizes the other way, for more loss on volatility than gain.  See this article for an example:

http://www.investopedia.com/articles/mutualfund/07/leveraged_etfs.asp

Then note that, unlike an unleveraged index fund, a leveraged one must trade in options or other derivatives to accomplish its stated goal, and thus may (and, it appears, does indeed) vary from the "theoretical" target of 2x return or loss. 

See these articles for discussion and look up the actual performance of the Ultra Pro version, which significantly underperformed the 2x goal in a positive year.  The "short" version, which is designed to go the opposite way of the index, performed even worse, losing money even when the index was losing and it should be theoretically gaining.

http://www.investopedia.com/articles/exchangetradedfunds/09/broken-leveraged-etfs.asp

You could probably run the track record of these funds so far with their much higher fee drag and see if they still beat the index over long terms. 

In short, these leveraged funds are designed for people who trade daily. 

[MrSal]

I've read (not just here, but just about everywhere) about how to achieve 7% average investment returns, after inflation, by investing in the stock market (SPY if you're a fan of the S&P 500, QQQ if you're a fan of the Dow, or VTI/VTSMX if you're a fan of the entire stock market). The key being that the market on the whole returns an average of 7% per year, after inflation, but each individual stock may or may not beat that average. So investing in broad based funds gives you the best chance of hitting that average (with low expense ratios to boot).

So why not use leveraged ETFs to accomplish a better return?

Obviously, leveraged ETFs are not for everyone. Granted. SPY, for example, isn't for everyone. Even the stock market itself isn't for everyone. But if you get past those issues, and we come to the conclusion that you will invest in the stock market, long term, and you will invest in broad based ETFs or mutual funds, why ignore the leveraged ETFs?

A few givens:
1. You save $X per paycheck.
2. You put that savings into a market account (IRA or Scottrade Account, or similar device).
3. You use that market account to purchase broad based ETFs or mutual funds.
4. You don't worry about the increase or decrease in the market, you just keep your head down and keep saving and investing, knowing over the long term you'll return a 7% average if you invest long enough.
5. Keeping that mentality, you determine SPY or QQQ is the best investment for you.

If that's the case, why not use UPRO, or UDOW, triple your earnings?

Yes, it may go down, but you've already decided that you'll stick it out and keep investing anyway. So why not live on the larger upside of a 3x leveraged ETF?

I hear everyone talk about buying SPY, QQQ, or VTI, but you don't hear about UPRO or UDOW. Other than larger risk, why not?

No.

Absolutely not.

I'd you don't know how leverage works in ETFs it works differently.

You will be Torned to shreds in medium/long term due to volatility decay.

Since I am on a phone, explaining the whole thing is hard.

But the reason sis simply due to the fact that these ETFs mirror the returns on a daily basis and ate reset for the next day.

You can guess what happens when allied with a bit of volatility, even when the original unleveraged ETF is back at the original point,  hence break-even, the leveraged ETF will be  50% or more lower. It will Always be behind due to this decay.

Trust me I worked as a trader for a hedge fund for 7 years and I've been trading in markets for 18.

I use them myself but only for very short term trades

[MrSal]

Due to the asymmetry present on negative returns and the returns needed to get back to break even, (a 50% looss needs 100% gain... A 75% loss needs 300%) you can guess what happens.

You can even use SPY returns as example and see what happens with double or triple daily leverage using data from SPY.  once I'm on the computer I can crunch the numbers so you can check it out

[specialkayme]

The compounding actually penalizes the other way, for more loss on volatility than gain.

Thank you for the articles. Very helpful. However, the specific events they point to don't match up to the overall history of the 2x ETF they pointed to.

Take a look at SSO (2x) vs SPY (1x).
At the height of the precrash market, SPY was worth 155.85. It crashed down to a low of 68.92. A loss of 55.7%.
During the same time period, SSO had a height of the precrash market of 49.77. It crashed down to a low of 8.66. A loss of 82%. (less than 2x SPY).
SPY was able to regain it's precrash high around 3/13. It's now worth 311% of it's crash low. 37% of it's precrash high.
SSO was able to regain it's precrash high around 12/13 (took almost a year longer, no doubt). It's now worth 806% of it's crash low (way more than 2x gains). 40% of it's precrash high (way less than 2x gains, but still more gains).

It still appears to me, based on history, that the downside risk is less (you lost 82% in the crash, as opposed to 2x 55.7%) with more upside potential (if you bought both at the crash low, you'd be up another 495%; if you bought both on the precrash high you'd still be better off (although marginally) than holding on to the SPY stock).

I'm not saying the 2x bear, the 2x bull, or the market itself won't go out of wack, as it records 2x of each days (not month or yearly) gains or losses. But overall it still appears to work out.

Unless I don't see something. Which is very possible.

[Jack]

Leveraged ETFs are only suitable for day-trading speculation, for complicated mathematical reasons related to their property of "resetting" daily. They do not actually track the index over periods of longer than one day! If you don't understand it -- and you don't -- then it's not suitable for you to invest in.

If you want leverage that bad, even trading options or using a margin account would be a less-bad plan, although I certainly don't recommend either.

The closest thing to reasonable leverage would be to invest in a normal (1x) ETF while not being debt-free (e.g. while holding a mortgage).

[secondcor521]

Can you explain to me how you came to the conclusion the risk/reward ratio takes a signifcant turn for the worse?

Who said you can't do both?

So, where did my math go wrong? I know it must have somewhere, or everyone would be talking about leveraged ETFs. . . .

Question 1:  First, logically, it is more expensive to manage leveraged ETFs instead of regular stock mutual funds.  This increases costs.  Second, logically, the only way to actually leverage is by borrowing, which increases costs and risks.  Although I don't understand the details, these funds don't actually deliver perfectly leveraged results as others have pointed out.  So you're actually paying more two different ways, increasing your risks, and not increasing your returns.

Question 2:  Nobody, and in fact anybody can do both.  I just decided for myself that, again, on a risk/reward basis, saving more was better for me than risking beyond 100%.

Question 3:  I don't know; see the others' comments in the thread.

I think it's always valuable to ask "Why not X?" but I also think it's wise to fully understand why others think something is not a good idea so that you can execute your strategy on a fully informed basis.

Good luck!

[specialkayme]

If you don't understand it -- and you don't -- then it's not suitable for you to invest in.

Oh trust me. I understand them. I understand they reset daily. I understand they are risky. I understand how they achieve a 2x or 3x return (conceptually, obviously I don't know what futures and derivatives they purchase). I also understand that a 2x ETF doesn't necessarily mean you'll walk away from a long term investment that is 2x the original benchmark.

However, what I do understand is if a benchmark ETF is rising, regularly, the leveraged ETF will also be rising.

I enjoy following the stock market. Not because I need to. Not because day trading is something I can realistically do. But because paper trading to find trends is interesting to me. I haven't found a "system" that always works. But it's fun to try anyway. Since January 1 I've been following the S&P500's gains on benchmark days based on when Stocktrader Almanac says it should be increasing or decreasing (which, by the way, YTD it's right exactly 50% of the time so far). As the benchmark, I wrote down the S&P500 on those same days, to determine if a buy and hold strategy would outperform Stocktrader's strategy. Right next to it, I keep track of UPRO's price to compare the same, on a buy and hold strategy.

YTD you know what I found? I'd break even (after paying commissions) on a regular day trading system with Stocktrader. I'd lose about .2% if I used the Stocktrader benchmarks in leveraged ETFs. Buy and hold on SPY I'd be up ~4.87%. If I bought a 3x leveraged ETF I'd be up 15.37%.

Is that always going to happen? No. But it did this year. It did between '09 and today.

I'm not talking about outperforming the market by 3x. I'm talking about outperforming the market overall. I'm not looking for 21% annual average. I'm looking for 7% annual average.

So tell me where I'm wrong here, mathematically. Seriously, I'd like to know, other than "it resets daily" or "you don't even understand, so just don't."

[MrSal]

The compounding actually penalizes the other way, for more loss on volatility than gain.

Thank you for the articles. Very helpful. However, the specific events they point to don't match up to the overall history of the 2x ETF they pointed to.

Take a look at SSO (2x) vs SPY (1x).
At the height of the precrash market, SPY was worth 155.85. It crashed down to a low of 68.92. A loss of 55.7%.
During the same time period, SSO had a height of the precrash market of 49.77. It crashed down to a low of 8.66. A loss of 82%. (less than 2x SPY).
SPY was able to regain it's precrash high around 3/13. It's now worth 311% of it's crash low. 37% of it's precrash high.
SSO was able to regain it's precrash high around 12/13 (took almost a year longer, no doubt). It's now worth 806% of it's crash low (way more than 2x gains). 40% of it's precrash high (way less than 2x gains, but still more gains).

It still appears to me, based on history, that the downside risk is less (you lost 82% in the crash, as opposed to 2x 55.7%) with more upside potential (if you bought both at the crash low, you'd be up another 495%; if you bought both on the precrash high you'd still be better off (although marginally) than holding on to the SPY stock).

I'm not saying the 2x bear, the 2x bull, or the market itself won't go out of wack, as it records 2x of each days (not month or yearly) gains or losses. But overall it still appears to work out.

Unless I don't see something. Which is very possible.

Sure 806% from the crash low... but how about from the point they both reached breakeven?

For someone that bought in 2007 and rode the wave until today?

SPY has a gain of around +70% (including dividends) vs peak of 2007 ... SSO has a return of +35% and that is after enduring a 85% loss!!

That is a 3% year return after enduring a 85% loss! And you say the risk reward ratio on SSO is atractive?

SSO is good for short term movements especially if the market goes in one direction trending very nicely. Central banks have helped these past 6 years to turn the market without volatility and be a one way street almost.

MAR Ratio ... Sharpe etc is completely ridiculous on this one.

[MrSal]

Example of 1 year of data:

[MrSal]

if you want to use SSO or other leveraged one, sure it is valid AS LONG it is for trading.

I use it myself on my 401k since I cannot use a margin account on my 401k and IRAs I use it in a small part of my portfolio, but only when there is a VERY consistent trend and volastility is very low.

If you want to get higher than market returns then a nice portfolio of a few asset classes, a trend overlay and a bit of leverage like 50% or so should be enough with a VERY BIG risk/reward propositon.

Like a max DDown of -15/-20% vs returns around +15%

[specialkayme]

If you want to get higher than market returns then a nice portfolio of a few asset classes, a trend overlay and a bit of leverage like 50% or so should be enough with a VERY BIG risk/reward propositon.

Like a max DDown of -15/-20% vs returns around +15%

And you lost me.

Can you explain what you mean?