I was reading this article about S&P 500 short ETFs:
https://www.zacks.com/stock/news/176217/4-ways-to-short-the-sampp-500-with-etfs
and I was confused when it said: "investors should note that such products are suitable only for short-term traders as these are rebalanced on a daily basis". What is ETF rebalancing exactly? and why does it make these ETFs unsuitable for long-term short trading?
The problem with daily-rebalanced "inverse" or "leveraged" ETFs is that since they rebalance every day, you can lose money even if your guess as to the market's direction is correct.
Quoting from FINRA'S guide as to why these are a bad idea:
How can this apparent breakdown between longer term index returns and ETF returns happen? Here’s a hypothetical example: let’s say that on Day 1, an index starts with a value of 100 and a leveraged ETF that seeks to double the return of the index starts at $100. If the index drops by 10 points on Day 1, it has a 10 percent loss and a resulting value of 90. Assuming it achieved its stated objective, the leveraged ETF would therefore drop 20 percent on that day and have an ending value of $80. On Day 2, if the index rises 10 percent, the index value increases to 99. For the ETF, its value for Day 2 would rise by 20 percent, which means the ETF would have a value of $96. On both days, the leveraged ETF did exactly what it was supposed to do—it produced daily returns that were two times the daily index returns. But let’s look at the results over the 2 day period: the index lost 1 percent (it fell from 100 to 99) while the 2x leveraged ETF lost 4 percent (it fell from $100 to $96). That means that over the two day period, the ETF's negative returns were 4 times as much as the two-day return of the index instead of 2 times the return.
That example is for "just" leveraging 2x in the same direction. Inverse funds have the same kind of issue. An example from Bogleheads Wiki page on these kinds of funds says that over 12/31/2007 to 12/31/2010,
- The Vanguard 500 index fund lost $846.
- ProShares Ultra S&P (SSO), the 2X ETF for the same index lost. But it didn't lose just twice as much as VFINX, it lost over four times as much-- $4,050.
- ProShares UltraShort S&P (SDS), the 2X short ETF (-2X, gains if benchmark drops) which should have earned a profit, lost even more--$4,595.
The funds do exactly what they say on any given day. But any losses get "locked in" each day. While normally a 50% loss needs a 100% gain to get back to a starting point, a fund like this needs more than a 100% gain to get back to its starting point. The result of these funds across multiple days doesn't match the index it's matching over those several days, and you won't make money over the long term. Do look at the further examples at the links I've referenced above, or do your own research into the performance of these funds during time periods both when the market is going up and going down.
Also refer to these related and/or duplicate questions:
Fund rebalancing typically refers to changing the investment mix to stay within the guidelines of the mutual fund objective. For example, lets say a fund is supposed to have at least 20% in bonds. Because of a dramatic increase in stock price and decrease in bond values it finds itself with only 19.9% in bonds at the end of the trading day. The fund manager would sell sufficient equities to reduce its equity holdings and buy more bonds.
Rebalancing is not always preferential because it could cause capital gain distribution, typically once per year, without selling the fund. And really any trading within the fun could do the same.
In the case you cite the verbiage is confusing. Often times I wonder if the author knows less then the reader. It might also be a bit of a rush to get the article out, and the author did not write correctly.
I agree that the ETFs cited are suitable for short term traders. However, that is because, traditionaly, the market has increased in value over the long term. If you bet it will go down over the long term, you are almost certain to lose money.
Like you, I cannot figure out how rebalancing makes this suitable only for short term traders. If the ETFs distribute capital gains events much more frequently then once per year, that is worth mentioning, but does not provide a case for short versus long term traders.
Secondly, I don't think these funds are doing true rebalancing. They might change investments daily for the most likely profitable outcome, but that really isn't rebalancing.
It seems the author is confused.
If you want to make a profit from long term trading (whatever "long term" means for you), the best strategy is to let the good performers in your portfolio run, and cull the bad ones.
Of course that strategy is hard to follow, unless you have the perfect foresight to know exactly how long your best performing investments will continue to outperform the market, but markets don't always follow the assumption that perfect information is available to all participants, and hence "momentum" has a real-world effect on prices, whether or not some theorists have chosen to ignore it.
But a fixed strategy of "daily rebalancing" does exactly the opposite of the above - it continuously reduces the holdings of good performers and increases the holdings of bad.
If this type of rebalancing is done more frequently than the constituents of benchmark index are adjusted, it is very likely to underperform the index in the long term.
Other issues in a "real world" market are the impact of increased dealing costs on smaller parcels of securities, and the buy/sell spreads incurred in the daily rebalancing trades. If the market is up and down 1% on alternate days with no long tern trend, quite likely the fund will be repeatedly buying and selling small parcels of the same stocks to do its daily balancing.
The problem with leveraged etfs is that volatility causes a parasitic drain on your fund. Even if an index stays the same value, a leveraged version of that index will decrease in value over time if there is any volatility at all. This is a consequence of basic algebra.
If any investment goes up by a certain percentage, x, one day, then goes down by the same percentage the next day, it will end up lower than where it started. This is a consequence of the mathematical fact that
(1+x)*(1-x) = 1 + x - x - x^2 = 1 - x^2,
which is less than 1 because of the term -x^2.
If the investment stayed the same, it actually went down some lesser percentage, y, on the second day, where y is exactly the number so that
(1+x)*(1-y) = 1 + x - y - xy = 1
But now if you double the percentages x -> 2x and y->2y, (through leverage) the investment actually decreases in value, because
(1+2x)*(1-2y) = 1 + 2x - 2y - 4xy
= 1 - 2xy
which is less than 1 because of the term -2xy. Here, we used the fact that x - y - xy = 0, which follows from subtracting 1 from the from both sides of the previous equation.
For example, here is a table of a fund that keeps going up and down, but ultimately stays the same, compared to a 2x leveraged version.
Day | Original fund | % change original | 2x leveraged fund
1 | 100 | | 100
2 | 105 | +5% | 110
3 | 100 | -4.76% | 99.52
4 | 105 | +5% | 109.48
5 | 100 | -4.76% | 99.05
6 | 105 | +5% | 108.95
7 | 100 | -4.76% | 98.58
You can see that the leveraged fund goes down in value over time, even though the original fund stays the same.
You can still make money if the original index goes up, but you can easily lose all of your gains from a bull market if, at a future time, the original index simply stays the same for a while (or worse, if it decreases).
In addition to the excellent answers here I might suggest a reason for investing in leveraged funds and the original purpose for their existence. Lets say you run a mutual fund that is supposed to track the performance of the S&P 500.
If you have cash inflows and outflows from your fund due to people investing and selling shares of your fund you may have periods where not all funds are invested appropriately because some of the funds are in cash. Lets say 98% of your funds are invested in the securities that reflect the stocks in the S&P 500. You will will miss matching the S&P 500 because you have 2% not invested in some money market account.
If you take 1/3 of the cash balance and invest in a triple leveraged fund or take 1/2 of the funds and invest in a double leveraged fund you will more accurately track the index to which your fund is supposed to track. I am not sure what percentage mutual fund owners keep in cash but this is one use that I know these ETFs are used for.
The difference over time that compounding effects have on leveraged funds is called Beta Slippage. There are many fine articles explaining it at you can find one located at this link.
