Post New

Hi All,

I have been playing with putting together a Cash Secured Puts portfolio based on Cohen's book using the ROE_CashLTDebt screen that Jim posted a while back as an MI version that would do slightly better than the SP500.

The basic screen steps are:
VL Timeliness 1-5 (timeliness exists) - 1175 stocks pass this
Sort by ROE Latest QTR (high to low), take top 33% - 388 stocks pass this
Sort passing stocks by Cash-LTDebt, take top 45 stocks

Ran this today and it generates the following list of tickers sorted by Cash-LTDebt:
GOOG, MSFT, TSLA, ELV, CSCO, HUM, ACN, RNR, MOH, COST, APO, NKE, NVDA, TJX, ROST, EOG, AMP, MELI, NVR, CPRT, EQH, FTNT, ALNY, SPNT, NYCB, TXN, SEIC, BBY, PBF, WST, NTAP, MLI, EW, ADBE, IDCC, EME, CFR, VSH, ENPH, BKNG, SEDG, NSP, ACLS, DKS, EWBC

After this, it gets slightly more complicated. Using this list of 45 stocks, I have a python script that goes through TDAmeritrade and pulls the PUT data out for each stock (first three put options OTM for the given date). For each option, I have bid/ask, target price (70% between bid and ask), %OTM, the premium (target price - strike + current stock price), the premium % (premium/strike), the hedge % (%OTM + 0.75*premium%), the corresponding strike of the SP500 hedge (current price * (1-hedge%)). The hedge and premium formula are all from Cohen's book.

Using this data, I create a list of descending premium%, and take the first 6 unique stocks, and then create a portfolio of puts and hedges. For example, using a portfolio size of $500K, and expiration of 19-Jan-2024, and todays numbers:

PRICE     DESCRIPTION                STRIKE  BID    ASK    TARGET  CASH REQ'd  OTM %  PREMIUM  CONTRACTS  REVENUE
273.2914  TSLA Jan 19 2024 270 Put   270     32.85  33     32.895  237.105     1.20%  13.87%   4           $13,158.00
41.02     PBF Jan 19 2024 40 Put     40      4.7    5      4.79    35.21       2.49%  13.60%   24          $11,496.00
435.985   NVDA Jan 19 2024 430 Put   430     48.55  48.75  48.61   381.39      1.37%  12.75%   2           $9,722.00
1142.27   MELI Jan 19 2024 1120 Put  1120    118.1  122.5  119.42  1000.58     1.95%  11.94%   1           $11,942.00
205.25    ALNY Jan 19 2024 195 Put   195     16.4   18.7   17.09   177.91      4.99%  9.61%    5           $8,545.00
120.58    EOG Jan 19 2024 118.2 Put  118.2   8.7    8.9    8.76    109.44      1.97%  8.00%    8           $7,008.00

This gives an average OTM% of 2.33% and an average preimum% of 11.63%. Thus using the Cohen formula, the SP500 Hedge strike would be 395, and thus 14 SP500 Put contracts would hedge the portfolio. Thus on a 500K cash, you would collect $61871 premium, but pay out $7133 for the hedge (11% of your potential profits), for a net of $54738 (or 10.3% on your cash pile, 20.6% annualized). Note that you can construct a table of what the loss including the hedge would be if the entire portfolio (and SP500) fell by a corresponding %. The worst case scenario there is a 15% loss across the board, resulting in a 0.03% loss.

Obviously, the problem isn't that everything drops together the exact same amount, but that your chosen stocks drop and the market does not. Also, I seem to recall Jim at some point saying that given that you will almost never have to buy all of the stocks in the set, you can actually only hold half of the required cash, which would of course increase the % by double. True disasters are hedged (the only thing that goes up in a down market is correlation), but you are only holding 6 positions, thus any one of them can crater on you without the broad market doing so.

Running this for various expiration dates, I get the following aggregate results:

DATE        OTM     PREMIUM  HEDGE   NET     ANNUALIZED
1/19/2024   2.33%   11.63%   11.53%  10.29%  20.57%
3/15/2024   13.68%  12.68%   6.95%   11.79%  17.91%
6/21/2024   2.71%   18.15%   10.70%  16.21%  17.23%
9/20/2024   4.80%   12.35%   22.09%  9.62%   7.99%
12/20/2024  5.15%   8.94%    40.34%  5.34%   3.67%
1/17/2025   4.43%   22.67%   12.55%  19.83%  12.63%

There is some judgement involved, but it would seem to me that the 21-Jun-24 expiration gives a nice tradeoff of high annualized return without a too crazy hedge percentage, though the Jan-25 doesn't look to bad either.

Looking forward to hearing your thoughts on this...and ask me if you need any additional details.

-Gabriel

I have done this both personally, and for a corporate client. (I used to manage money for a fee)
I was much more conservative for the corporate client: more mechanical, more diversified, more hedged, fewer hand-chosen positions.

If choosing the stocks mechanically, you really do want as much diversification as you can manage.
As you note, the only real risk of capital loss is that your stocks drop and your chosen index doesn't.
So, to the extent you can, pick your stocks for index correlation, not necessarily highest premium.

Taking those comments together, puts on Tesla as one position out of six seems pretty risky.


I have a very old rule of leverage, too, which could be worth considering. Or not.
FIRST choose something predictable, without worrying too much about the rate of return. A positive return even in the relatively depressing scenarios.
THEN add the leverage.
There isn't much overlap between the choices with the juicy upside and the choices for which it makes sense to add leverage : )
In this case, the individual positions aren't particularly predictable, but the hedged slate is not bad, which is why some leverage is quite liveable.


In real life, I make most of my returns in a pretty risky way that kinda follows that rule of thumb: leveraged positions on Berkshire Hathaway, primarily via calls.
There is essentially no compounding of the profits, as I live from my portfolio.
I calculate my average dollars at risk as what I'd lose on a given day if the stock went permanently to zero, averaged across all days since my first position.
For every $1000 of my average dollars at risk on Berkshire in the last 21.85 years, I've made an average profit of $266/year on Berkshire positions.
Wild variation year to year, and some long dry spells that required gritted teeth, but a good average.

Jim

Hi Jim,

Thanks for weighing in. First off, the warning of:
There isn't much overlap between the choices with the juicy upside and the choices for which it makes sense to add leverage : )
is always a good one. Leverage in this case is going to be relatively small in that it is just how much of the cash pile to keep on hand.

I'm not sure exactly how to optimize for index correlation. Given that I have 45 stocks that are pretty well correlated with index (though obviously beleived to have a bit better prospects due to the high ROE and Cash positions). None are expected to go under any time soon. Of the 45 stocks, maybe 25 have options available. I can't sell puts on all of them as a single contract is still 100 shares, so the cash required to backstop the position gets too big for the portfolio.

Any ideas (from anyone) on how to optimize for index correlation?

I'm thinking of implementing this in a small corner of my portfolio, just to see what it does. The other think to think about is that while this is a long strategy, it is harvesting time premium, which tends to be higher with less time to expiration (2-3 months to go seems to give the largest premium evaporation). Perhaps this is better run on a shorter time scale. Thoughts?

--G

PS: As far as the BRK calls, I assume you are talking about the purchasing of DITM calls when the Price/Peak Book is within the lower range, and selling the calls and buying back the stock when the Price/Peak Book is higher. Though that should probably be on the BRK board.

Any ideas (from anyone) on how to optimize for index correlation?

Oddly enough, using a bit of leverage might help.
If you are writing $100k face value of puts against $100k of cash, you can only do so many.
If you are writing (say) $160k face value of puts against $100k of cash, you can do 60% more positions.

In my view, the main thing about diversification is not to look mainly at historical price correlation.
Rather, think about firms that have different kinds of moving parts.
Interest rate sensitive or not?
A lot of cash or a lot of debt?
Very US focus, focus elsewhere, or very international mix?
Tech fashionability or fuddy duddy?
Big China exposure?
High P/E or low P/E?
Very recession dependent/cyclical?
Dividend payer or not?
Luxury goods or discount goods?
End consumer or business to business?

The reason is that when something bad happens that is NOT transient (the thing you're worried about), it tends to hit a group of firms that are similar in some way.

Of course, the simplest starting point is just to make sure you have at least one position in every sector.

There is no rule that says you have to enter your long positions and your hedges at precisely the same moment.
You're not an investment banker with some box-checking risk committee staring at you.
For example, imagine a given 3 week period with one very high index/happy day that you buy your insurance puts, and one crashy terrible day that you write your single-name puts.
This can make a huge difference to the annual return.

The main thing is that it's a fair bit of work. You want to watch the portfolio.
You'll often find that a position is doing well after a little while: you might make 75% of the maximum possible profit in only 30-50% of the elapsed time.
It's easy to calculate your maximum annualized remaining rate of return on capital at risk.
If it's below a certain number, close the position and replace it with something else. (either immediately, or on the next panicky day).
I used 8%/year rate for a while, though you can pick any number you like.

After a few years I stopped doing the mechanical insurance puts, and it was mainly a long-only portfolio.
I spent more and more time picking the firms I wanted to write repeated puts against, so I had more and more comfort being long my picks.
I did do other hedging, but on a very different cycle: entered only when the market timing entrails seemed to say it was a good time to put them on.
I made very steady returns for a while, but if I look at my returns on the hedging only, I lost a fortune or two.
First rule of hedging: NEVER look at the return of just one side! Always look at the totality.
Because one side is always a huge loser!

Jim

After a few years I stopped doing the mechanical insurance puts, and it was mainly a long-only portfolio.
I spent more and more time picking the firms I wanted to write repeated puts against, so I had more and more comfort being long my picks.
I did do other hedging, but on a very different cycle: entered only when the market timing entrails seemed to say it was a good time to put them on.
I made very steady returns for a while, but if I look at my returns on the hedging only, I lost a fortune or two.

Yes, I agree, when I was doing a hedged set of cash secured puts (using a set of MI screens as the underlying), I found that the hedges were always losers, even when the market declined substantially. At that point, even though the hedges were worth something, the event was usually quite transitory and by the time I closed the position, the hedges had bounced back down in price and didn't help (in one case actually losing money).

I'm trying to set this up mechanically so it is a "rinse, recycle, repeat" kind of operation, with the built-in discipline of repeating every few months. I agree with calculating the "return to go" and closing the position early once that has dropped to less attractive levels. In terms of not putting the hedge on immediately, the worry is that if you are unlucky, then the right day to put the hedges on simply doesn't come--and now you are unhedged.

--G

Post New

Mechanical Investing FAQ

---

Contact Shrewd'm
Contact the developer of these message boards.