Dividend capture with covered calls

[Rehoboth]

Quote from bjw:

no he didn't

there wouldn't have been any time value to collect when opening this position a day before the dividend. there's no free lunch here, really. option prices adjust as much as they can without doing less than intrinsic, because everybody knows the div is coming. if you get assigned, which you likely will, you pay costs of getting into position in the first place (cost if the options aren't that liquid could be quite heavy) + commission.

the trick of betting on people not exercising is widely know. marketmakers, with a fraction of our comm costs, were writing & buying heavily already. yesterday's volume in the nov 47 was over 5k in contracts with an oi of 270 or so. the trick is simple: the mm exercise their own 5k, which is spread over all 5270 contracts by a lottery; even if people wouldn't have exercised here, which they did, the posters here would still have been extremely(!) unlikely to not get assigned and probably the marketmakers would have taken it all anyway.

Like I said, I did them for a credit, so yes I did make money. A whopping 9 cents after commish.

Not getting assigned would have produced a decent winner, but like bjw said, it would have been stupid not to exercise. When I used to do this in the 90's and 2k's is was around a 75 percent rate, but i can only imagine the efficiency right now.

 

[madbrain]

bjw,

Quote from bjw:
if someone wants to get out of their position, they should just sell it , not exercise.

I see, makes sense now.

the only reason when people should exercise is the day before the div when div > time value (edit: and obviously upon expiration itself)

OK. As I understand, there is auto-exercise at expiration for calls that expire in the money. Provided you have the cash available.

to strenghten my point, i did look up there were no exercises in the jan-13 30 yesterday. you would definitely not have been assigned. if you want me to look up other series let me know; they will 100% confirm what i'm saying.

Thanks.

It sounds like the game might work better when the time value and the dividend are close.

Since you offered to look at another series, here is one
SPY and Dec 30 $106 call . I chose this to get a time value as close to the projected dividend as possible.

SPY is $121.96 . Bid is $16.56 / Ask $16.91. I know these are after-hour prices, but I think they are still accurate.

Buy/write would cost $121.96 - $16.56 = $105.40 . So, time value is $0.60 .

The last SPY dividend was $0.625. Next dividend will be around Dec 19, before the expiration of the call on Dec 30.

So, here we have time value = $0.60 < dividend 0.625.
How likely is it that this one will be exercised early ?

This provides 13% downside protection. Max net return would be only 0.625/105.4 = 0.59% at expiration, with a duration of 42 days. 5.1% annualized.

If called early, the return would be 0.60 / 105.40 = 0.57%. Let's say it's called on dec 19, the day before dividend. Duration = 31 days. 6.7% annualized.

Of course, there is always the risk SPY goes below $105.4 before dec 19. In which case you lose.

 

[Rehoboth]

Quote from madbrain:

bjw,



I see, makes sense now.



OK. As I understand, there is auto-exercise at expiration for calls that expire in the money. Provided you have the cash available.



Thanks.

It sounds like the game might work better when the time value and the dividend are close.

Since you offered to look at another series, here is one
SPY and Dec 30 $106 call . I chose this to get a time value as close to the projected dividend as possible.

SPY is $121.96 . Bid is $16.56 / Ask $16.91. I know these are after-hour prices, but I think they are still accurate.

Buy/write would cost $121.96 - $16.56 = $105.40 . So, time value is $0.60 .

The last SPY dividend was $0.625. Next dividend will be around Dec 19, before the expiration of the call on Dec 30.

So, here we have time value = $0.60 < dividend 0.625.
How likely is it that this one will be exercised early ?

This provides 13% downside protection. Max net return would be only 0.625/105.4 = 0.59% at expiration, with a duration of 42 days. 5.1% annualized.

If called early, the return would be 0.60 / 105.40 = 0.57%. Let's say it's called on dec 19, the day before dividend. Duration = 31 days. 6.7% annualized.

Of course, there is always the risk SPY goes below $105.4 before dec 19. In which case you lose.

You cant predict where the spy is going to be the day before the ex-date so you cant predict what the time value is going to be for the spy. So if the time value is greater then the div, then you will probably capture the div, but you will more then likely have a whole different problem on you hands in that the spy would have moved down much closer to your strike.

 

[bjw]

Quote from madbrain:


Since you offered to look at another series, here is one
SPY and Dec 30 $106 call . I chose this to get a time value as close to the projected dividend as possible.

SPY is $121.96 . Bid is $16.56 / Ask $16.91. I know these are after-hour prices, but I think they are still accurate.

Buy/write would cost $121.96 - $16.56 = $105.40 . So, time value is $0.60 .

The last SPY dividend was $0.625. Next dividend will be around Dec 19, before the expiration of the call on Dec 30.

So, here we have time value = $0.60 < dividend 0.625.
How likely is it that this one will be exercised early ?

This provides 13% downside protection. Max net return would be only 0.625/105.4 = 0.59% at expiration, with a duration of 42 days. 5.1% annualized.

If called early, the return would be 0.60 / 105.40 = 0.57%. Let's say it's called on dec 19, the day before dividend. Duration = 31 days. 6.7% annualized.

Of course, there is always the risk SPY goes below $105.4 before dec 19. In which case you lose.

as the previous poster said, to determine the chance of it being exercised you'd have to look at the time value the day before dividend.
assuming time value is still 60 ct the day before dividend and the dividend is 66 cent, it would be pretty much a coin toss if you're asigned i'd think. every investor is facing different comm cost, exercise cost, is in a different tax situation, etc. for some it would be rational to exercise, for some it won't be (and some just aren't rational at all ofcourse).

but really, it's time to take a step back and think about what you're doing. in the beginning you tried to come up with a strategy to capture div, by now all you're doing is trying to cash in on time value. this time value isn't there for nothing, there's a small chance spy will drop below 106. assuming option prices are efficient, the chance of it dropping is reflected by the current price. if you believe spy-options are overvalued because of recent volatility, fine, but this would be a position trade, and nothing more. the div is pretty much irrelevant. you could essentially even be better off just selling the put at $1.10 - $1.16 and pray SPY stays over 106.

 

[madbrain]

Quote from madbrain:
This provides 13% downside protection. Max net return would be only 0.625/105.4 = 0.59% at expiration, with a duration of 42 days. 5.1% annualized.

Actually, I miscalculated this case. At expiration, if SPY is above $106, I get called at $106 and thus collect the 0.60 of time value.
Thus the net profit would be 0.625+0.60 = $1.225 / $105.4 = 1.16% over 42 days or 10.1% annualized.

 

[madbrain]

Rehoboth,

Quote from Rehoboth:

You cant predict where the spy is going to be the day before the ex-date so you cant predict what the time value is going to be for the spy. So if the time value is greater then the div, then you will probably capture the div, but you will more then likely have a whole different problem on you hands in that the spy would have moved down much closer to your strike.

Edit :
It seems to me that the time value could have gone up for other reasons than the underlying price going down. Volatility could just go up, too Possibly enough to counter the fact that 70% of the time to expiration elapsed. In which case time value may go up, and I would collect the dividend.

 

[madbrain]

Quote from bjw:

as the previous poster said, to determine the chance of it being exercised you'd have to look at the time value the day before dividend.
assuming time value is still 60 ct the day before dividend and the dividend is 66 cent, it would be pretty much a coin toss if you're asigned i'd think.

OK.

but really, it's time to take a step back and think about what you're doing. in the beginning you tried to come up with a strategy to capture div, by now all you're doing is trying to cash in on time value. this time value isn't there for nothing, there's a small chance spy will drop below 106. assuming option prices are efficient, the chance of it dropping is reflected by the current price. if you believe spy-options are overvalued because of recent volatility, fine, but this would be a position trade, and nothing more. the div is pretty much irrelevant. you could essentially even be better off just selling the put at $1.10 - $1.16 and pray SPY stays over 106.

Yes, the put may have greater profit, but does not have the same likelihood of being called early. As it turns out, holding to expiration is better (see corrected calculation above) as you would collect both the dividend and time value.

 

[bjw]

Quote from madbrain:

OK.



Yes, the put may have greater profit, but does not have the same likelihood of being called early. As it turns out, holding to expiration is better (see corrected calculation above) as you would collect both the dividend and time value.

which would sort of equal what you'd get for the put, and probably less when you take div taxes into account (i'm just guessing, i haven't done the math).
but you can't calculate this as a strategy that yields say 10% annualized, because, again unless you believe option prices in the series you are trading are wrong (which i haven't heard you say), one out of many trades will go wrong and wipe out all profits. the div itself offers no edge here, assigned or not assigned, there could be another edge (or much more likely there won't be, especially in something as liquid as SPY) but this thread isn't about that.

 

[madbrain]

bjw,

Quote from bjw:

which would sort of equal what you'd get for the put, and probably less when you take div taxes into account (i'm just guessing, i haven't done the math).
but you can't calculate this as a strategy that yields say 10% annualized, because, again unless you believe option prices in the series you are trading are wrong (which i haven't heard you say), one out of many trades will go wrong and wipe out all profits.

No, I don't think the whole series of option prices is wrong.

the div itself offers no edge here, assigned or not assigned, there could be another edge (or much more likely there won't be, especially in something as liquid as SPY) but this thread isn't about that.

True. I wouldn't have another hedge, as in this case it would be guaranteed to wipe profit.

This weekend I was looking at a couple of other issues which might be more interesting for this type of play.

JNK has a high yield. However, what's fairly interesting is that the dividend is paid monthly, not quarterly. And thus, the individual dividend is always very small. This might compare more favorably against the time value. Ie. the monthly dividend might be more likely to be lower than an option's time value.
It seems to me this might significantly reduce the odds of getting called each month. Of course it depends what time value one is able to find on a long-term, deep in-the-money call.

By the way, what's up with the strange strike prices on JNK options ?
Like strike 29.39 through 49.39 for Jan 2012 and 34.39 through 44.39 for Jan 2013 ?

 

[Rationalize]

Hey -- maybe read the Baird's "Option Market Making", or Natenberg's "Option Volatility & Pricing". You may find that everything's actually priced-in. Better info in the books than on the internet(s). * Some market makers get their rookies to read Natenberg.

p.s. The JNK strikes are probably an adjustment due to a corporate action / capital adjustment. But, I'm not going to look into it.