langlands wrote: ↑Sun Oct 18, 2020 10:51 pm
thwang99 wrote: ↑Sun Oct 18, 2020 8:11 am
Shorting instead of going long has been talked about. I did a quick and dirty test using PV, and shorting has higher CAGR while having lower max drawdown and a significantly higher Sharpe ratio. What reasons are there not to use the shorting strategy? I picked TBT since it is easily shorted on IB (I actually did short SQQQ/TBT successfully for a few months).
(I know the ratio of 60/40 isn't optimized, but it's more the comparison we care about here)
TQQQ/UBT long: https://www.portfoliovisualizer.com/bac ... tion2_1=40
SQQQ/TBT short: https://www.portfoliovisualizer.com/bac ... ion2_1=-40
This thread is so long that inevitably we begin running around in circles. The validity of shorting leveraged ETFs to "harvest" the volatility decay has already been discussed at length about 20 pages ago.
You say that "Shorting instead of going long has been talked about." Did you take a look at viewtopic.php?p=5399207#p5399207
and the subsequent discussion?
I will simply reiterate what I said then that shorting volatility decay is not a free lunch and that shorting SQQQ and rebalancing on a daily basis is the same as going long TQQQ (excluding expenses, borrowing costs).
+1, there is no free lunch.
The problem is these funds didn't exist during the 2000 and 2008 crashes, so folks can't see how dangerous they are. But we can crudely estimate in PV on a monthly basis for 2008-2009, using a 55/45 allocation to monthly leveraged SPY and TLT.
https://www.portfoliovisualizer.com/bac ... ion3_2=135
What this suggests is that shorting SPXS/TMV would have been much better (-3.83% over the two years) than longing SPXL/TMF (-35.13%). But only if the position was held through a 104% loss
. At the bottom of the market, the long strategy would have been down 64% from its original value, while the short strategy would have been completely wiped out. Not to mention that as the market goes down and the short exposure goes up, so do the borrowing costs. Another way to look at this is if TQQQ goes down 10% for five days, you've lost 41% of your investment. If SQQQ goes up 10% for five, you've lost 61%. So during a steady decline, the short position has the potential to lose far more, however because the exposure is increasing rather than decreasing, it will also recover faster. That recovery is an example of the anti-volatility decay benefit, but the tradeoff is more risk. PV doesn't capture that risk since it uses monthly numbers.
Now, maybe the investor has a lot of other assets and is okay with being wiped out, but since maintenance margin is something like 90% for leveraged ETFs, at the worst point she would have only been able to keep a little over 50% of her portfolio in this. And bearing in mind the rest of the portfolio likely would be down too, that's maybe more like 25% of her portfolio that could have safely be held just prior to 2008. If she held more than that, the broker starts closing positions and that kills the excess returns.
In... short... this strategy is interesting, but limited in practice. I am employing it for fun, but I am keeping my exposure to 3% of my overall portfolio for safety (so when the value increases, I'll short more to bring it up to 3%, when it goes down I'll close positions to bring it back to 3%). What this looks like in a typical rising market is that the majority of the gains get moved into regular HFEA holdings rather than reinvested, so I won't see anything close to the difference in returns seen in PV backtests. In order to fully realize the difference in CAGR you need to fully reinvest in the short strategy and let it compound.
There are a few unique properties that are nice though, apart from volatility decay working in your favor, mostly taxation related. 1) Little rebalancing is needed with a fixed allocation, since if stocks go up and bonds go down, the short exposure will do the opposite, so it is largely self-balancing. 2) Rebalancing is tax free or even a tax loss harvest because you only ever close positions that lost value, otherwise you're just shorting more. 3) Borrowing fees can be deducted from taxes, so aren't as much of a drag as they might seem. 4) I maintain -100% funds +200% cash, which provides extra liquidity in a pinch, i.e. I can effectively borrow against myself; more aggressive investors might deploy that 200% somewhere other than cash the whole time. 5) When the market goes down and I need to close positions to restore the -100% funds +200% cash ratio, it's a tax loss.