4. When NOT to hedge
Derivatives 5: Hedges and strategies
This is the level's important module, and the one least often taught. The previous three explain how three hedges work and what they cost. This one explains when the right answer to the question "how do I hedge" is that you should not be hedging. Before going on it is worth spelling it out: this is education about mechanisms and prices, not advice or a recommendation to trade derivatives. The aim is that you should know how to work out the cost of each trade-off; the decision is yours.
First: when the accumulated cost exceeds the risk covered
We already have the sum. The two 450 puts cost 3,000 dollars a year on the 100,000 portfolio. Ten years of continuous protection leave a compounded drag of around 26%. The S&P 500's maximum drawdown over recent decades has ranged between 20% and 55% depending on the episode, and that put only pays whatever exceeds 10%.
Write it as an actuary would: you pay a certain, recurring premium against an uncertain, occasional claim. If your horizon is long, the premium is charged every year and the claim does not come every year. That is why permanent protection has turned out, over most of the long periods measured, dearer than the falls it protected against. There is a serious public debate about this between the defenders of systematic tail hedging and those who maintain that other ways of mitigating risk work out better. Neither side disputes the arithmetic of the premium; they dispute whether the comfort it buys is worth its price.
Second: when the instrument does not correspond to what you want to cover
This is basis risk, and it is the most expensive mistake because it looks as though you are covered. Suppose your 100,000-dollar portfolio is not an S&P 500 ETF but twelve mid-cap technology stocks, and you hedge it with index puts because they are the liquid, cheap ones. The correction arrives:
- The index falls 12%. Your 450 put, with a 10% excess, pays 2 points: around 2,000 dollars gross, which does not even cover the 3,000 of premium.
- Your portfolio falls 34%, because mid-cap technology stocks have a far higher beta than the index and carry their own sector punishment.
You have lost 34,000 dollars and your hedge, net of premium, has cost you 1,000. Worse still is the mirror case: the index rises while your sector falls. There the hedge loses money at the same time as the portfolio. Hedging with a correlated but different instrument is not hedging; it is adding a second bet whose result depends on the difference between two things nobody controls.
Third: when you are hedging a position that does not actually worry you
It is worth separating two things that get confused all the time: volatility and damage. Volatility is your portfolio falling 30% in March. Damage is having to sell in March. If your horizon is twenty years, you make regular contributions and you are not going to need that money, a 30% fall gives you an unpleasant few weeks but does no permanent economic damage. Paying 3% a year for twenty years to avoid an unpleasant few weeks is a legitimate decision, but it is worth knowing that you are buying emotional calm and not protection of your wealth, and that the tariff is 3% a year.
The opposite case is where hedging does make sense: when the date you need the money is fixed and close. If you are going to pay the deposit on a house in eleven months' time out of that portfolio, a 30% fall destroys the plan and it makes no difference that the market recovers in two years. There the insurance covers real damage bounded in time, and its price is compared against something tangible.
Fourth: when the hedge only serves to avoid cutting a position that is too big
The most common and the worst. Somebody has 40,000 dollars in a single share out of a 100,000 portfolio, knows it is too much, and instead of selling buys puts. He has turned a sizing problem, which is solved once, into a recurring cost that is never solved. The premium on a put over an individual stock rarely comes in below 5% a year; on a 40,000 position that is around 2,000 dollars every year, indefinitely, as long as the position is still too big.
And here is the alternative almost nobody puts on the table: reducing the exposure is the cheapest hedge there is. Compare the two routes on our 100,000 portfolio:
- Buy the 450 puts. Cost of 3,000 dollars a year. Effective floor at 87,000. The portfolio is still 100% exposed on the upside, less the premium.
- Sell 25,000 dollars' worth and leave it in Treasury bills at 4%. Cost of the transaction: the commission and the capital gains tax, once. From then on you take in 1,000 dollars a year instead of paying 3,000. A 30% fall in the market becomes one of 22.5% for your wealth.
The difference between the two routes is 4,000 dollars a year of cash flow, 4% of the portfolio, in favor of the boring route. What you lose by reducing is expected return through less exposure, and that is an honest, frictionless trade. What you lose by hedging is certain money every year.
One final note on timing. People want to hedge once the fire has already started, which is when implied volatility is through the roof and the policy costs three times as much. On March 16, 2020 the VIX closed above 82, its highest close on record. Buying protection that day was buying insurance with the house already ablaze.