Hammerling makes a good point.
My objection is to TRADING options as a strategy.
If you just buy put options (or some simple bearish spread) with a long time to expiry just as an insurance, and you understand the terms of that insurance, and you don’t try to change that insurance by trading it often, that makes sense. Your mathematical expectancy will still be negative, but it’s the same for any insurance you buy (that’s why insurance companies make money), and who cares? It’s for your peace of mind.
On the other hand, if Yuval’s system has a very low beta un down markets, he doesn’t need insurance.
This is very interesting indeed. I’m not going to get negative beta going long only. I can get low beta and high alpha, but I’m still expecting to lose money in the next recession. So if I want to take out some insurance with 2 to 5 percent of my portfolio, I guess I have four options: buy puts on SPY, like Kurtis suggests; buy the SH ETF; buy puts on companies that I think are going to do especially poorly (e.g. Tesla); or invest in a bond ETF. Or, as Yoram suggests, just rely on my low beta system and forego insurance. My question for Kurtis is what makes this strategy superior to SH–or buying long-term puts on overvalued stocks?
Yuvaltaylor,
I am actually not suggesting buying puts on the SPY. That was just as an example for simplicity sake. I would develop an S&P 500 short system where there is short alpha. Take 5 - 10 stocks and buy puts on those. Be wary of stocks that you think are over-valued. Every single time I have been burned - from Netflix to healthcare stocks developing therapies. Stocks without cashflows or limited earnings have their price based on future expectations. Often, these stocks can defy gravity as they are priced on sentiment. Try to find stocks which you think will crash in a recession - like a diamond jewelry store or premium brands of some sort.
The big difference between buying cheap put options and something like the SH is this…
Suppose you are wrong and the market goes up 10% per year - every year. With your put options - you are down 2% (assuming you can get full coverage of your long position with 2%). If you are 100% hedged with SH and the market goes up 10% - you lose 10%. You are only earning long alpha now. A market neutral portfolio.
So when options are cheap - you only give up 2% of your total upside for a potential 50 - 100% downside protection. With actual shorting, you are giving up all of the markets returns - whether that be 2% or 20%.
Puts only protect you from market going down BELOW THE STRIKE PRICE (I assume you hold them as insurance untill close to expiration, not trading them, which has it’s aformentioned problems), so you still lose money if markets are stable, or if they go down a little (and note if they go down, then up, you still lose unless you decide to sell your puts thus giving up your insurance). Of course, the higher the strike price, the more likely the insurance will pay off, but the premium will be higher. That’s why its hard to make a profit buying options. Of course, insurance is not meant to make a profit, just to give protection in case…
Anyway if you decide to buy puts, note to insure your beta position, not your total postion value.
Haven’t read all of the discussion but I agree with the general tone of hemmerling’s mesage.
In general, conventional shorts are much less risky than buying puts. The reason is that you sell a stock short and buy, for instance, SPY with the proceeds. This reduces your risk.
With puts, you need to get a lot of things right: the time frame, the strike price and the price you pay for the put. And their value decays over time.
On average (in long term) you lose money when you buy options (especially equity index options).
The premium (IV - RV) is compensation for peace of your mind.
On the other hand, currently IV in most market is extremely cheap (eg. VIX is below 10) so maybe it is not bad move buy cheap insurance on SPY if you anticipate sharp move in near time.
I don`t think this is true. When you short stocks directly your risk is infinite, while buying puts limits your risk to the money invested. Assuming you have a short port that is profitable, i can imagine that the profitability goes up using in the money long term options because you limit your downside and have more than 100% upside. I do this myself right now and sell puts on stocks in the long port to offset premium paid. I think this is much safer than purely shorting stocks and i have the advantage that i can run a concentrateted short portfolio, while conventionally you are forced to diversify the short port.
Synthetic shorts, created by selling an ATM call using those proceeds to buy an ATM put, have two advantages:
- Leverage
- You don’t owe dividends (like you would if you shorted an idividual stock)
If the stock fell in price, then you would gain through the purchased puts, but if it increased in price, then you would lose from the written calls. The potential profits and the potential losses are roughly equal to what they would be if you were purely short selling the stock.
Dividends and cost to borrow are factored into option prices. So the call is cheaper than the put and you pay the dividend and cost to borrow that way. Except when you find a mispriced pair of options. Even interest that you get by selling stock is factored into the option prices. I think it only makes sense to create a synthetic short if you are not able to short the stock otherwise.
First thanks for bringing this up. I have to revise myself, shorting with synthetic shorts can make a lot of sense. I played around with my short model and a lot of option implementations last week and i think i have an implementation now that is close to what you suggested. I decide which options to buy/sell now on a case by case basis.
- If implied volatility for the stock/options is low (<25%) i just buy straight at the money or in the money puts (OTM puts are often more expensive).
- If IV is higher (25-50%) i create a synthetic short with an embedded stop loss by selling an ATM call and buying an OTM call (20-30% above the current price) (so this is a bear call spread) and an ATM put option.
- For really high IV situations put options alone are to expensive and there is no put/call parity because of borrow costs, so i only found bear call spreads and sometimes bear put spreads as profitable implementations, but the returns on these are much lower because the reward:risk is often 1:2 and the probability of success is often not enough to offset that. (But since these are stocks with high borrow costs, maybe the probabilties for success are higher than the averaged backtest results.)
My spreadsheet backtest said that this implementation of the port improves the short-model-performance of ~+7% to ~+20%. Does anybody trade a short portfolio with options and can confirm what i found?