Why writing covered calls is (usually) smarter than writing puts.
First, a note on risk
The difference is that writing puts requires much less capital up front and so you dramatically increase your ROI with an equivalent risk/reward profile. Sometimes it makes me wonder why we bother selling CCs at all.
Reducing the amount of up-front capital does not increase your ROI with an equivalent risk/reward profile. Just ask anyone who has used margin. (Margin poses two problems: required interest payments and increased risk).
Think of it this way. I'm going to offer you the following bet: Give me $1, and flip an honest coin. If it's heads, I'll give you $1.50. If it's tails, I'll give you $.75. You can do this once a year. The "expected return" of this strategy, as typically calculated, is (.50 + -.25)/2 = 12.5%.
What do you think your long-term return will be if you follow the above strategy? (write down your answer).
You have another option. You can make the above bet with only 50% down. So you give me $1, and flip an honest coin. If it's heads, I'll give you $2. If it's tails, I'll give you $.50. So now instead of gaining or losing 50% or 25%, respectively, you gain or lose 100% or 50%, respectively. The "expected return" of this strategy is (1 + -.5) / 2 = 25%.
What do you think your long-term return will be if you follow the above strategy? (write down your answer).
If you think that the second strategy gives you a higher return, albeit with higher risk, you're wrong. The first strategy will give you a long-run annualized return of 6.066%, and the second strategy will give you a long-run annualized return of 0%. If you had put less than 50% down, you would have a negative return.
That's because the real way to calculate an expected long-run ROI isn't by finding the probability-weighted arithmetic average of the returns, but by the probability-weighted geometric average.
For instance, if you place the first bet for 100 years, you'd win about 50 times, and lose about 50 times, for a result of .75^50 * 1.5^50 = $361, or 6.066% annualized.
In the long-run, your return is just .75^.5 * 1.5^.5 = 1.06066
Likewise, the long-run return of the second strategy is .5^.5 * 2^.5 = 1
If you paid only 25% down, the long-run return would be .25^.5 * 2.5^.5 = 0.79 (that's -21% annualized!)
I should point out that sometimes leverage does in fact increase long-run returns. For instance, if the original bet paid out 1.25 on heads, and .875 on tails, the return would be a mere 4.6%, so leveraging it to simulate the first bet above would be wise. For more on the intuition behind this, read up on the Kelly Criterion.
The point is, leveraging your bets doesn't always increase your returns, and excessive leverage can be absolutely fatal. I would venture that leverage is addictive--more investors use too much than too little. A glass of wine a day may be good for your heart, but if your family has a history of alcoholism, best is probably to avoid the stuff altogether.
So the ability to leverage your returns in a put-writing strategy isn't necessarily an advantage. Besides, you can leverage CC's too (by buying the underlying stock on margin).
By the way, calculating expected ROI as an arithmetic, not geometric, return makes more sense if you're heavily diversified. The above analysis is biased by my belief that successful investing requires focus on a small number of stocks, so heavy diversification is not an option. Furthermore, it's very tricky to separate diversifiable risk from non-diversifiable risk--e.g. any tech investor who was heavily leveraged in 2000 got annihilated, regardless of how "diversified" they thought they were.
Why writing CC's is (usually) smarter than writing puts
First, let me establish that writing a CC is similar to writing a cash-secured put (i.e. a put with no leverage... 100% down).
Supposed you have a stock at 13.5. Let's say your considering either a buy-write CC at 15, or writing a put at 15. Call the time premium of the put P(p), and the premium of the call P(c).
If you write the CC, your initial outlay (in cash, if it's a buy-write, or in opportunity cost if you already own the stock) is 13.5 - P(c). What's the payoff? Well, let's look at the potential results depending on the price at the end of the month.
8 100 shares
9 100 shares
10 100 shares
11 100 shares
12 100 shares
13 100 shares
14 100 shares
15 $1500
16 $1500
etc.
If you instead write the put, your initial outlay is 13.5 - P(p). The payoffs are:
8 100 shares
9 100 shares
10 100 shares
11 100 shares
12 100 shares
13 100 shares
14 100 shares
15 $1500*
16 $1500*
etc.
* -- it's possible that the payoff will be higher than $1500, e.g. if the put is exercised, and then the stock goes up. This is because puts are sometimes exercised before the expiration date.
So as you can see, the payoffs are very similar. Unsurprisingly, the commission costs & tax consequences can be a bit different, though.
There are a couple reasons why the payoffs from the put-writing strategy can be a bit better, though. First, the put-writing strategy lets you collect interest on your initial outlay--about .5% a month, if the risk-free return is 6%. Second, see the asterisk above.
But these are both pretty minor. We'd expect the P(c) and the P(p) to be virtually identical, since they provide for virtually identical returns. Note that the high bid/ask spreads obviate arbitrage, so it's possible that they won't be exactly the same.
And in fact, they are pretty close, although generally not identical. In scanning through the option prices for RMBS, NTAP and MSFT, it appears that that P(c) is usually substantially higher than P(p). This may be because the market has fairly bullish expectations.
Anyway, this implies that as of now, writing covered calls is probably smarter, since you get higher premiums for the same payoff graph. If the market had more bearish expectations, it's possible that P(p) would exceed P(c), and so put-writing would be smarter (at least in a large, tax-exempt account). |
