There has not been much discussion in a while about shorting and models around them. Is shorting even worth the effort? Do the challenges around estimating availability and borrowing costs pose to great an issue?
Issues of availability and borrowing costs, while interesting, are not the most important. The most important issue is that shorting dramatically increases your chances of going broke. You’d have to be wildly confident in your model to assume you’d effectively mitigated that risk.
I am speaking to a short model as part of an overall long/short portfolio with a large enough number of positions (40+) to mitigate stock specific risk.
I have some experience with shorting (in a portfolio of around 50 positions as part of a long-short portfolio).
I work with Interactive brokers. The availability is good and borrowing costs are OK for almost all stocks except for many small cap (Borrowing cost can easily bee shown as a data item in IB Trader Workstation).
Borrowing costs are unavailable in Portfolio123, so you have to estimate. With experience, this is possible knowing the type of stocks your system wants to short.
Assuming your system shorts mainly small caps, you will have to go down the rankings to find shorting stocks with reasonable borrowing costs.
So, in back-testing test to see how your short system is doing with lower ranked stocks (and of course input the appropriate borrowing cost).
The stock specific risk isn’t the risk I’m referring to. I’m referring to model specification risk. If you pick 40 stocks in your long/short portfolio based on 1 model, and that model is temporarily wrong and you experience a 50% drawdown in your long/short portfolio, you are now bankrupt. If you are long only using no margin, and you experience a 50% drawdown due to your model being temporarily wrong, you still have half your money.
In my opinion shorting is a bad, bad, bad idea for individual investors and you should not ever do it. Just because it has worked in the past does not make it a good idea. Institutional investors have some protection because their exposure is limited to the funds they invested as LPs in whatever hedge fund or CTA is doing the shorting, but as an individual you do not have that same benefit if the shorting is happening in your personal account.
This has been said before on these boards and I believe it to be true. The best hedge is cash. Yes you can come up with some great back test shorting or hedging. But in real time with my money the best hedge is cash. You will eventually shoot yourself in the foot trying to make up some extra ground shorting. I don’t think it is worth it. Too many moving parts.
Shorting is definitely a good idea as long as you have some alpha. You don’t even need to profit on your shorts to increase your portfolio risk-adjusted return, since shorts will be inversely correlated to the rest of your portfolio.
Sometimes - you can buy and short stocks in the same industry, and trade just the spread between the two stocks. I have done this very successfully.
Only true if you know with absolute certainty that your alpha will not turn negative tomorrow. If you are that certain, you are probably more confident than you should be.
You are confusing volatility and risk. Volatility is what happened, risk is what could have happened (likelihood x impact). Volatility can be measured ex post, risk cannot. By way of example, if you go out and get terribly drunk and then drive home without crashing your car and killing anyone, that does not make driving drunk just as safe as driving sober.
I know I’m being a real buzzkill about this because the prospect of having a perfect looking exponential return curve that sails straight through crashes and recessions is super exciting. But the kinds of biases on display here can be ruinous and so I feel obligated to point that out. That is super boring (I know) so I apologize for that.
Great comment !!
Thank You.
“the prospect of having a perfect looking exponential return curve that sails straight through crashes and recessions is super exciting. But the kinds of biases on display here can be ruinous”
… In 2008, I had a mkt neutral portfolio…half long and half short. I was short a bunch of finance and bank stuff. Seems like that would be great. My longs started getting killed, no surprise. But my shorts barely fell and my losses got surprisingly big. I was shocked. A temporary short sell ban hurt me plus funds were massively deleveraging their portfolios… covering shorts and selling longs…and I obviously owned the same types of value stocks.
I still short all the time. But you gotta be careful!
The risk of shorting can be managed just like any other risk. I find shorting dramatically reduces my risk and I’ve been doing this for decades.
Thanks all for the inputs. I think overall holding a portion of the portfolio in cash or intermediate treasuries is better choice than trying to manage a short book. There seems to be alpha available in shorting small caps, but the cost to borrow and availability would be a major headwind.
If you are looking to hedge, then I would suggest shorting via ETFs. It avoids many of the pitfalls and costs associated with specific stocks. There are more to choose from than in the past and you can do some correlation work based on what your longs look like. There is no free lunch of course, as you do have the underlying costs of the derivatives used on the inverse side. I agree 100% with what others have said in that it isn’t easy and it can be dangerous.
I promise that the following question is not sarcasm. Would you be willing to explain how taking a position with unlimited downside reduces your risk? I understand that in practice it has lowered your volatility and perhaps raised your realized returns. But how has it reduced the likelihood that you could have sustained an unfavorable outcome?