Major Factors Affecting Call-Writing Returns

Stock, Strike Price, Expiration, Volatility, Interest Rates, Strategy, Institutional

Course: [ OPTIONS FOR VOLATILE MARKETS : Chapter 3: The Basics of Covered Call Writing ]

Returns from covered writing will depend very much on how you implement the strategy your choice of stock, strike price, expiration, and your follow-up program (rolling).

Major Factors Affecting Call-Writing Returns

Returns from covered writing will depend very much on how you implement the strategy—your choice of stock, strike price, expiration, and your follow-up program (rolling). Above all, covered call writing is an equity strategy, and its results will be reflective of the performance of the underlying equities in one’s portfolio.

Stock Selection

If you are picking stocks specifically for a covered writing portfolio, you may find that you will gravitate toward a somewhat different set of criteria from that employed to select stocks for outright purchase, whether as long-term holds or as short-term trades. Instead of picking stocks for long-term appreciation potential, you will more likely focus on the shorter-term scenario, where the “shorter term” will coincide with the duration of the option you write. That will generally lean you more toward technical analysis than fundamental analysis as your decision mechanism.

A common trap in call writing is to look for stocks with the largest option premiums under the assumption that fatter option premiums will generate higher returns on the same amount invested. They do generate higher returns, but only when they succeed. It’s much the same logic as betting on the horse at 10-1 odds instead of the horse with 2-1 odds. The 10-1 horse will pay five times as much, but should only win one-fifth as often. Fatter option premiums exist on higher-volatility (i.e., higher beta) stocks because the stocks tend to move more in both directions. So, while you receive more option premium, you have a commensurately higher risk on the stock itself. A better approach is to determine whether the options on a particular stock are abnormally high or low, and that is discussed in the Volatility section later in this chapter.

It is also important to remember that higher call premiums cannot necessarily be taken as a signal that the market feels the stock has a greater chance of moving up than down. As we mentioned previously, both call and put options will expand in price when volatility in a stock jumps. Bad news on a stock may drop the price and create a surge in demand for puts, but in that situation, both calls and puts will expand due to higher implied volatility. Therefore, if you are focusing only on the calls for covered writing and you see that the percentage return from a covered call write on the stock has jumped recently, it might have been due to greater expectation of a drop in price than a rise. A classic example of this would be BP stock after the Deepwater Horizon rig explosion in April 2010. In the aftermath of the explosion, the stock dropped precipitously and implied volatility skyrocketed. If you were using an online filter to screen for covered writes with the highest potential returns, BP would have surely been near the top of the list at that point. But the reason it was there was because holders were desperate for put protection in case the disaster caused additional harm or sent the company into bankruptcy. The resulting implied volatility made the stock look like an enormously attractive covered call write at that point in time—but not necessarily because the market was overly bullish on the stock. (One way to give yourself a sanity check on a stock with high option implied volatility is to see if the volume of trading activity is higher on the call side or the put side.)

There is no right answer for everyone regarding stock selection, whether writing calls or not. As with buying stocks without writing calls, you are always well-advised to diversify, regardless of your method of stock selection. A big boost to covered call writers is the introduction of ETFs on market sectors, broad indexes, and even commodities, upon which calls are available. Covered call writing on these ETFs may reduce or even eliminate entirely your need to diversify your holdings, and the call-writing strategy works exactly the same on ETFs as on individual stocks. Chapter 8 is devoted to a discussion of options on ETFs.

Strike Price

The strike price with the greatest amount of time value is always the one closest to the current price of the stock. Consequently, the farther away the strike price is from the current stock price, the lower the static return (return unchanged) for the position. Table 3.2 illustrates this point.

Remember that this is not the only factor you should consider when deciding which strike to write. Lower strike prices will provide greater downside protection, while higher strike prices offer greater potential gains from the stock. You should, however, understand how these trade-offs affect your static return.

You can see from the table that the difference between writing the Sep 35 call and the Sep 40 call is 5.4 percentage points. Remember that this is only a two-month return. On an annualized basis, the difference between writing the near-the-money call and the next higher strike price is more than 30 percentage points in return. Of course, this is the static return and therefore ignores appreciation in the stock, but it gives you a perspective on how much return from option premium you give up when you write a higher strike price in the hope that the stock will move up. The higher you go in strike price, the greater the potential return from a covered write, but the lower the contribution from option premium and the more uncertain the overall return.

If we take Table 2.1 from Chapter 2, which showed how time value varied with strike price, we can look at the impact that has on covered call returns.

TABLE 3.2 Call-Writing Returns at Different Strike Prices             

Price of XYZ stock = 50

Volatility = 20%

Days to expiration = 60

Interest rate = .5%

Strike Price

Option Price

Time Value

Annualized Return if Exercised

Annualized Return if Unchanged

40

10.04

.04

.5 %

.5 %

42.5

7.57

.07

.9 %

.9 %

45

5.21

.21

2.6 %

2.6 %

47.5

3.15

.65

7.9 %

7.9 %

50

1.64

1.64

20.0 %

20.0 %

52.5

.71

.71

39.0 %

8.6 %

55

.26

.26

64.0 %

3.2 %

57.5

.08

.08

92.2 %

1.0 %

60

.02

.02

121.9 %

.2 %

Clearly, the selection of strike price will vary results extensively. But you will not know which strike is optimal when you initiate the position. Therefore, it is necessary to go with a strike price that fits your risk tolerance or upside expectation each time you write. Writing OTM calls such as the 57.5 or 60 strike calls in Table 3.2 provides an exceptional return when the stock makes a 7 to 10 percent move in 60 days, but provides subpar returns when the stock does nothing or declines during that period. Unless there is reason to believe the stock may have a significant drop or rise in the next two months, the strikes between 47.5 and 52.5 provide the best combination of attractive return and safety.

Expiration

The selection of expiration month will also affect the expected returns of your covered writing program, because options do not lose time value (decay) at the same rate throughout their life. The differing rate of decay is important and can have a significant effect on the returns from the strategy. It is also why we emphasize that while covered writing can certainly be an ongoing approach, each individual write is a relatively short-term strategy.

TABLE 3.3 Percentage Return from Different Durations

Price of XYZ stock = 50

Volatility = 20%

Interest rate = .5%

 

Expiration

Time Premium

Premium per day

% Return Unchanged (Annualized)

30 days

1.15

.038

28.0 %

60 days

1.64

.027

20.0%

90 days

2.01

.022

16.3 %

120 days

2.33

.019

14.2 %

360 days

4.07

.011

8.1 %

In Chapter 2, we showed how time value diminishes slowly at first and then accelerates closer to expiration. The faster time value declines in an option, the better it is for the writer, and the writer will garner more time premium by writing one month options for three months in a row than by writing a single three-month option to start. This means that option writers garner higher annualized returns by writing shorter duration options, and the difference can be considerable. Table 3.3 shows the difference in annualized returns by writing options of different durations. For purposes of comparison, one has to assume all other factors remain equal and that the ETF remains at exactly the same price for a year. While this is obviously not a reflection of reality, we can at least isolate the time value aspect to demonstrate the difference between durations. The new weekly options (available only on a dozen or so stocks and ETFs at the present time) offer a unique opportunity for ultra short term covered writes. This technique will be discussed further in Chapter 5.

There are expected trade-offs to using the shorter duration options in covered writes. The writer needs to take follow-up action more often during the year and transaction costs are accordingly higher. One concern about short duration writing is that the calls have less time premium and therefore offer less downside protection than longer duration options. It is true that you take in more total premium by using a more distant expiration month, but you don’t necessarily gain much in near-term protection unless the stock falls considerably.

TABLE 3.4 Price Protection from Long-Dated Calls       

Price of XYZ stock = 50

Volatility = 20%

Interest rate = .5%

 

Expiration

Protection from 1-Point

Drop

Protection from

3-Point

Drop

Protection from

5-Point

Drop

30 days

.44

.51

1.11

60 days

.47

.66

1.46

90 days

.48

.67

1.55

120 days

.50

.77

1.79

360 days

.53

.93

2.23

That’s because more distant options assume wider price swings over time and do not vary as much with price moves in the stock. In option parlance, they have lower delta. Using the same example again, Table 3.3 shows how much protection would be provided by the various expiration months available in scenarios in which the stock drops in the first month to $32, $30, and $25.

Table 3.4 shows the differing time values and returns for call options with the same strike price but different expirations. The strike used is 50, because it is ATM and thus all the option premium is time value. Returns are annualized for comparison purposes, since the time periods involved are all different.

Table 3.4 shows that which expiration provides the most protection depends on how much the stock actually drops. In general, for better overall returns from covered writing, you will want to write a relatively close option most of the time. But if you want protection from the possibility of a significant drop in price, a more distant call may be warranted. We recommend covered writers generally use durations of one to three months and then adjust to fit personal preferences.

Volatility

As shown in Chapter 2, the time value in option premiums varies directly with volatility, and since stocks all have different inherent volatilities, the stock selection on your covered write takes volatility into account and will determine your potential return from that call write. But as we stated earlier in this chapter, one should not jump to the conclusion that writing calls on the most volatile stocks will necessarily provide the highest returns. While their premiums, and thus the potential returns, are indeed greater, the downside risk is proportionately greater as well. A better way to give yourself a statistical edge in call writing is by writing calls on stocks where the implied volatility (that forecast by the price of the call option) is greater than the actual volatility that is characteristic of that particular stock.

In other words, writing call options on higher volatility stocks doesn’t necessarily provide an edge, but writing options that tend to be overvalued, whatever the volatility of the stock, may improve your success. One must bear in mind, though, that determining when an option is overvalued is not an exact science. (If it were, the arbitrageurs would be all over it.) The Black- Scholes formula provides a guideline. Because it uses actual historic volatility, we can identify options that are trading above the Black-Scholes price and conclude that they are trading above historical volatility for that stock. But that, by itself, doesn’t mean they are necessarily overvalued. In reality, there is no way to know when options are truly overvalued or when they are perhaps more accurately reflecting future volatility.

An additional approach is not just to look for the highest premiums you can find relative to stock price, but to look for high implied volatility relative to previous implied volatility for that particular stock or ETF. This is discussed further in Chapter 5.

Interest Rates

Since option premiums are affected by interest rates, returns from covered writing will be as well. It takes a substantial move in rates, however, to affect premiums. (LEAPS options of one or more years in duration are much more sensitive.) Table 3.5 shows the effect of different interest rates on premiums and returns for a hypothetical two-month option.

Covered Writing as an Ongoing Strategy

The prevailing assumption in the early days of options (1970s and 1980s) was that writing calls on stock positions had merit only when specific situations existed, such as a concentrated position in low cost basis stock that a client wouldn’t sell, or speculative fever surrounding potential takeover candidates. Brokers and professional money managers had a difficult time justifying the sales of covered calls on stocks they liked for long term fundamentals as they assumed call writing would force them to accept lesser long term returns.

TABLE 3.5 Effect of Interest Rates on Option Returns

Stock Price = $30

Strike Price = 30

Days to Expiration = 57

Volatility = 0.30

Interest Rate

Premium

Return Unchanged

(Not Annualized

2.0%

1.46

4.9%

3.0%

1.49

5.0%

4.0%

1.51

5.0%

5.0%

1.53

5.1%

6.0%

1.56

5.2%

7.0%

1.58

5.3%

8.0%

1.60

5.0%

9.0%

1.63

5.4%

10.0%

1.65

5.5%

And when the fundamentals soured on a given stock, the tendency would be to simply sell it and buy a different one. No doubt, such sentiment may have been supported by the fact that much of the industry made their living on commissions.

The BuyWrite Index (BXM)

Meanwhile, institutional investors kept the flames of covered call writing alive and began using it frequently enough that they petitioned the CBOE to provide them with a covered writing index upon which they could benchmark their performance. The CBOE responded to their wishes in April 2002 with the creation of the CBOE BuyWrite Monthly IndexSM (BXM). The BXM tracks a hypothetical portfolio that is long the S&P 500 index (SPX) and short the one-month call option on that index at the nearest out-of-the-money strike price. As each option expires, the net gain or loss is registered, and the next month’s option is written. (For more information on the details of the BXM, visit the BXM micro site at www.CBOE.com.) Like most other major market indexes, the BXM is designed to emulate a passive (unmanaged) return against which portfolio managers can gauge their performance on actively managed accounts. In addition, perhaps even more importantly, the BXM allowed the investing world to see how a passive call-writing strategy performs over time with the buy and hold strategy on the underlying index itself. When originally introduced, the BXM has been calculated back to June 1988, when Standard & Poors began tracking the dividends on the S&P 500 Index. The returns of the S&P 500 Index (or more accurately, the S&P Total Return index which has the dividends reinvested) were then compared with those of a basic covered writing program (as represented by the BXM) over more than a decade prior to launching the BXM. The results created a stir among professional investment managers who spend their lives searching for ways to produce competitive returns with less risk.

As expected, the BXM outperformed the S&P index in years where the S&P was down or had single digit returns, and underperformed versus the S&P during years when the S&P performance was strongly positive. The BXM also lowered the volatility in monthly performance by about one-third over the S&P. The biggest news of all, however, was that the lower volatility did not come at the expense of long-term return, as people tended to expect. Instead, returns over the 13.5-year period came within 0.2 percentage points of the S&P 500 returns for that entire period (13.88 percent for BXM versus 14.07 percent for the S&P)—a remarkable showing for a strategy that lowers risk.

Bear in mind that these results provide a very broad benchmark of what a basic mechanical covered writing program can do compared with the buy- and-hold approach. Your portfolio is unlikely to be anywhere near as diverse as the S&P 500. Also, the BXM assumes a mechanically implemented program using one-month at-the-money calls only, without any rolling, and it doesn’t take into account transaction costs or taxes. Nevertheless, the BXM clearly demonstrates that although the returns from covered writing may be lower than those stocks that are capable of producing during a given period, they are commensurate with the returns stocks actually generate over the long term—with a substantial reduction in volatility. Thus, it provides an important affirmation of the long-term validity of covered writing.

Since 2002, the BXM has continued to perform in line with the prior 13 years, and the CBOE added a related index (BXY) that utilizes a slightly more out-of-the-money call each month. As one would expect, the BXY will perform better than BXM in up markets and not as well in down markets. A chart of the results from 1988 through 2010 for both BXM and BXY versus the S&P is shown in Figure 3.4. At this point in time, both the BXM and

FIGURE 3.4 Comparison of BXM, BXY, and SPTR


BXY are outperforming the S&P for the 22-year period, no doubt helped by the lackluster performance of the S&P during the last 10 years.

Summing Up Covered Writing

Chapter 4 discusses the various ways to implement covered call writing. But before we go there, we thought we’d summarize the basic pros and cons of the strategy.

Advantages

  • Lower volatility. Whatever the underlying instrument, covered call writing on it will lower the volatility. The BXM has been shown to have approximately one-third less month-to-month volatility on the S&P 500 Index when writing an ATM call option each month.
  • Similar long-term performance compared with buy-and-hold. The reduction in volatility (as shown for the BXM) does not reduce the long-term performance of a straight buy-and-hold strategy on the same underlying instrument.
  • Modest short-term downside protection. ATM or OTM covered writing provides modest downside protection in the short term while still allowing some upside potential. ITM call writing can provide greater short-term protection, but will allow very little, if any, upside.
  • Easily modified. Covered call writes are easily closed or modified once implemented by executing an offsetting transaction.
  • Generates income. Call writes generate cash flow on holdings that can be withdrawn, used to purchase other investments, or compounded.
  • Flexible use. Call writes can be used intermittently or as an ongoing strategy.
  • Used in more complex strategies. Call writing is often used in more complex strategies to help offset the price of other options (discussed further in Chapter 7).
  • Disadvantages
  • Limited upside. Upside potential is capped for the duration of the option, unless the option is closed or rolled.
  • Partial protection. Call writing only protects the downside risk of the underlying instrument to the extent of the option premium received.
  • Requires attention. Effective covered writing requires monitoring positions more closely and potentially executing follow-up transactions.
  • False security. Call writing can provide a false sense of security for those who do not embrace the fact that the underlying stock position still has substantial risk.
  • Introduces additional behavioral considerations. Call writing may result in a loss on the option sold. Even though the gain on the stock would be greater, taking a loss on the option is sometimes emotionally difficult. (This is discussed further in Chapter 4.) 



OPTIONS FOR VOLATILE MARKETS : Chapter 3: The Basics of Covered Call Writing : Tag: Options : Stock, Strike Price, Expiration, Volatility, Interest Rates, Strategy, Institutional - Major Factors Affecting Call-Writing Returns


Related Courses




Related Topics