3. What an options position really costs

Derivatives 9: Derivatives in a real portfolio

This module does something the course has been putting off: adding up. An options position has a theoretical payoff profile and it has a bill, and the bill does not appear in full anywhere. Part of it is charged as commission, part is paid inside the price without anyone noting it down, part arrives in the form of shares you did not want and part does not exist as a charge at all because it is money that has earned nothing.

Cost one: the spread, on every leg and in both directions

It is the largest of the four and the least visible, because it is not invoiced: it is paid by accepting a price worse than the mid.

The arithmetic is that of the Costs course on the easy track, with one aggravating factor specific to derivatives: it is paid per leg. Buying a call is one leg; a spread, two; a straddle, two; a one-by-two ratio, two legs of different sizes; a butterfly, three or four. And it is paid on opening and on closing. A spread opened and closed is four crossings of the spread, not one.

On a contract of 100 shares, a spread of 0.10 is 10 euros there and back. It looks like small change until it is multiplied by legs, by contracts and by months.

Cost two: commission, which goes per contract

Options commissions are nothing like share commissions. They are not usually a percentage of the amount, but a fixed sum per contract, to which market and clearing fees are added. The usual order of magnitude in the retail market sits between half a euro and two euros per contract, with per-order minimums at some intermediaries.

The consequence is that the cost scales with the number of contracts and legs, not with the money put up. Ten two-leg spreads are twenty contracts on opening and twenty on closing: forty contracts per cycle. The same exposure taken in shares would have been one order.

Cost three: exercise and assignment

There are three layers here, and the most expensive is the second:

• The exercise or assignment commission charged by the intermediary, of a similar order to a normal commission or somewhat larger.
• What arrives afterwards. If you are assigned on a short put over ten contracts, 1,000 shares you had not decided to buy appear in your account. Unwinding them costs the share commission and the underlying's spread, and doing it the same day means trading at the worst possible moment.
• The early assignment of American options, which level 2 introduced. A call sold on a share about to pay a dividend can be exercised the day before, leaving you short of shares and obliged to pay that dividend yourself.

Cost four: capital tied up as margin

This one appears on no statement because it is not a payment, but income that does not happen. Money blocked as margin is not invested in something else. If the risk-free rate is 3% and you have 19,000 euros tied up all year, the opportunity cost is 570 euros. If the intermediary pays interest on the balance, subtract whatever it pays; if it does not, the cost is the whole thing.

This cost is the one that forces you to measure the return of an options strategy on the capital committed and not on the premium. A strategy taking in 1,000 euros a month while tying up 19,000 does not yield a spectacular percentage on the premium: it yields on those 19,000, which is capital that is doing nothing else.

How this bill is reduced

• Fewer legs and less turnover. Every leg added and every cycle of the calendar is another spread and another commission.
• Limit orders near the mid, never at market, accepting that sometimes you will not trade at all when there is no price.
• Strikes and expiries with a real counterparty, following the previous module.
• Closing before expiry everything you do not want turning into shares.
• Counting margin as capital employed and calculating the return on it, not on the premium taken in.

Worked scenario: the annualized sum for a monthly strategy

Round figures. A portfolio of 100,000 euros. Every month ten put spreads on an index are sold, with a width of 20 points between strikes and a multiplier of 100, and they are closed or left to expire at the end of the month.

• Maximum risk per cycle: 20 points times 100 times 10 contracts, that is, 20,000 euros.
• Net premium collected: 1.00 point, that is 100 euros per spread and 1,000 euros a month. Annualized, 12,000 euros gross: 12% of the portfolio. That is the headline figure.

And now the full bill:

• Spread. Twenty live contracts, ten per leg, with a spread of 0.10 per contract there and back, that is, 10 euros per contract: 200 euros a month, 2,400 euros a year.
• Commissions. Forty contracts per cycle between opening and closing, at 1.20 euros including fees: 48 euros a month, 576 euros a year.
• Assignment. Suppose two of the twelve months end with the sold leg in the money and shares have to be unwound: of the order of 300 euros per episode between commissions and the underlying's spread, 600 euros a year.
• Capital tied up. Some 19,000 euros permanently blocked, at 3% risk-free: 570 euros a year.

Total bill: 4,146 euros a year. Against the 12,000 gross, costs take around 35% of the gross result. And measured against the portfolio, those 4,146 euros amount to an annual cost of 4.1%, against the 150 euros a low-cost index fund would cost per year on that same wealth. The comparison is not a moral judgment: it is the yardstick the strategy has to justify itself against, because that 4.1% is paid in the good years as well as the bad.

The worst is still missing, and it is not a cost: it is the result. A single month in which the index closes below the lower strike leaves a loss of 20,000 euros minus the 1,000 of premium, that is, 19,000 euros. More than two years of net result in one expiry, with the cost bill already paid in advance and non-refundable.