This is based on an article from big picture. If I long a stock, I have to invest capital to buy the stock. My opportunity cost is that capital tied up while invested in the stock. If I short a stock, I borrow the stock from someone else, and receive the proceeds from selling the stock. But the question is, what is my opportunity cost here? Besides having to pay dividends should the stock yields one, I’m not having any capital tied up at all. Are there any opportunity cost here so as to speak?
I struggled with this question when I first started doing shorting. Shorting SEEMS costless. It’s not. The trick is that you need to think about a short position in terms of the RISK to your capital, not how much it costs to enter the position. Once you learn that, you realize that you should be thinking about your long positions this way too. A short position ties up capital that you now allocate to cash just in case the position moves against you. Your broker will likely insist on having some cash or margin posted for this contingency. Example: Let’s assume you have an account with $100k of cash in it. You decide to short 1000 shares of ABC, now trading at $100 per share. You sell it after borrowing, so $100x1000 = $100k of cash enters your account at the time of shorting. We’ll ignore transaction costs and margin interest for now. You now have $200k in your account. $100k original capital + $100k of short proceeds. All is groovy as long as the position moves your way… you think “hey, this didn’t cost me anything to get into, it’s like free money, I like this shorting stuff.” Now, suppose the position moves against you. ABC goes up to $125. You panic and decide to cover. You pay 1000x$125 = $125k to buy your shares back. Your account value is now: $200k - 125k = $75k. It cost you nothing out of pocket to enter the position (excluding transaction costs), but it still cost you $25k over the course of the trade. — What’s the lesson here. The capital in your account doesn’t represent the maximum you can afford to *BUY* it represents the maximum you can afford to *LOSE*. In a long position, the two are more or less the same, because a long position is usually limited on the downside, and that’s why long-only folks can fairly comfortably get by by thinking about how much to buy. In the shorting world, there theoretically no limit to the downside, and there is a limit of 100% gain on the upside. Thus, what you want to do is figure out how far the position needs to move against you (in %) before you cover your short. THAT number times the nominal size of the short (price*shares) is the amount of your portfolio that you’ve really budgeted for the position. Same logic goes for a long. Decide how much the position needs to move against you before you run for cover. That percentage times the initial position size is the risk you have allocated to that position. If you prefer to work with volatilities and standard deviations, you can also use standard deviations to calculate risk allocations, using something like 2xSD or 3xSD to represent risk. Just remember that returns are not truly normal, and you will underestimate your true risk a fair amount when Black Swans come along. Normally, you want the sum of the risks for all positions in your portfolio to be some fraction of your total portfolio value - totals around 5-8% for total risk are typical. The mathematics of sizing positions according to their risk is a little trickier than just allocating percentages to different positions, but once you start doing it, you realize that it enables you to think of longs and shorts in the same framework, which is a big advantage.
Thanks bchad although a simpler thing to note would be that in the real world, a broker will require you to post margin against your short position even in excess of the short proceeds, which will themselves be restricted. So shorting does eat up your trading capital.