Subject: Re: An options strategy
IF the stock price fluctuations are random, log-normally distributed, AND the options are Black-Scholes priced, AND there is zero friction: no bid-ask spread on options prices, no fees for transactions, THEN the expected return from covered calls is precisely zero!
I would be interested in knowing if Jim agrees with that or at least finds it not implausible.

Another key requirement would be the academic assumption that the investor can borrow without limit at the policy rate, in a way that does not introduce risk.
For example, the leverage provided by in-the-money call options has no value to an investor if that's true. I can't borrow at the policy rate without limit or risk, so the assumption is false, so the calls definitely add a lot of value for me.

But yes, it's a good summary. If you assume that stock price movements are well characterized by random walks, as the options market generally does, then there is no advantage to using them. The options market is extremely efficient in that sense.

But as that assumption is manifestly not true--undervalued things are more likely to go up than down, and vice versa--the fact that the options market DOES make that random assumption means that they are indeed leaving "free money" on the table. Obviously an investor has to have very solid skill at identifying at least one asset which is clearly undervalued or overvalued, but that's far from impossible. In this case I can't say that Berkshire's price is going to move down soon, but I can say with strong conviction that "flattish" or "down" are both considerably more likely than "up" over the next several months. That's relatively obvious information for me, but that the options market isn't considering. Thus I can and do have an edge.

It should be noted that the shorter the time frame is, the closer to random price movements are, and the more accurate the assumptions in B-S are. At the scale of an hour, it's almost perfectly true and the calculator weenies are geniuses. And the reverse: the longer the time frame, the more preposterous the assumption can become. For example, the long-dated index put options that Berkshire wrote. They ultimately turned a tidy profit of over $4bn, which was pretty predictable statistically: the nominal value of an equity index will tend to rise over time, not follow a random walk.

You could press into service a quote meant for a very different purpose: what the wise man does in the beginning, the fool does in the end.

As another possible example: Say Berkshire is trading at $190 per B share two years from now. Would you be a buyer? I sure would. Let's say it's 99.99% likely to be a wildly profitable thing to do. Yet there are people out there today who are willing to pay you cash today for committing to doing what you would do anyway: they'll buy a put option you could write. Put another way, they are willing to pay money for the right to do something--sell Berkshire to you at $190 per B---that is 99.99% likely to be an unprofitable thing to do. This may seem really dumb on their part, and it is, taken as such. But may still make sense for them as a business venture, as they may be doing it across hundreds of stocks, and the assumptions become more true taken across many samples and it isn't worth their time to consider the outliers.

Separately, I believe there are some other corners of the options market which are "inefficient". But in a very much smaller way, so the built-in assumptions are not so unreasonable.

As mentioned, I don't think the "steamroller" analogy is at all apt in the case of writing calls against an existing Berkshire position when it is reasonably richly valued. In particular, technically there is zero chance of an incremental loss...only the risk of having a smaller gain than you might otherwise have had, and a 100% chance of reducing your maximum downside if the firm went bust the next day.

If you want to dismiss that by including the penalty of missing out on long term rises in the stock price after the time period of the option (because you sold and would not otherwise have done so), you're talking about the situation that you WANT a long term position. In that case, you would would presumably simply buy back the stock again. In that situation, you have to assess the odds that you would buy back at a price higher than the [strike+premium] at the moment you considered it richly valued enough to write a call. Pretty low chance, and if you did, the disadvantage would likely be modest (not a steamroller event).

e.g., if like me you think $410 per B is already getting a little rich and you take $25 premium for an at-the-money call, are you going to buy back in at a price above $425? Low odds, I would think. And if you did, you probably wouldn't do it much higher than that, so the disadvantage would be modest. In my own case, I think it's pretty darned likely I could wait a bit and get a re-entry lower than $400, a virtual certainty lower than $425. If so, then even if my call were exercised because the stock soars for a while I'd have made a tidy profit on the whole exercise.

Given the strong odds in my favour of doing it profitably once, it becomes a virtual certainty to be in my favour if done repeatedly over time. It's not picking up pennies in front of a steamroller, it's flipping a coin biased moderately in your favour, with no large downside outcomes. In your favour because you are taking into account the fact that the stock is rather richly valued when you enter the position, and the option counterparty isn't.

I don't write these calls much, as Berkshire isn't in my view sufficiently richly valued enough of the time to make it a good wager. Usually I'm at the other end of things, writing puts against things I consider to be either cheap or at most fairly valued. I've made an average of about $35 per contract over the years doing that. Not much, but it adds up.


Jim