Broader market volatility had been falling for a while, but one large sector held it off the floor: semiconductors. At the end of July, semiconductor volatility finally turned lower, and only then did the broader indices reach their floor. Chips have grown into a heavy weight in US and global markets, so this one sector's swings pulled the whole market.
What pushed chip volatility up
Semiconductor volatility rose through much of Q2 and July while nearly everything else fell. Two forces drove it. First, levered ETFs targeting the sector, both single names and the sector as a whole. Second, a large hedge fund holding concentrated, leveraged positions in the same area.
Both got their leverage the same way: total return swaps. A total return swap is an agreement between the fund or ETF and a bank, usually acting as the prime broker. Through these swaps the banks ended up warehousing risk from the ETFs and from the hedge fund at once, piling up exposure in the same names and in highly correlated names on their balance sheets. At some point the size got dangerous. Picture a chief risk officer seeing $50, $70, even $100 billion of exposure to one name or a cluster of names, and deciding the bank had to hedge against a sudden drop. That hedging is what drove volatility sharply higher across the sector, and dragged broader market volatility up with it, through May, June, and July.
Positioning, not fundamentals
Roughly 80% of what happened was positioning and 20% was market plumbing. The hedging need was very specific. Take a 2x levered ETF: the bank is only exposed if that name or index falls more than 50% in a single day. That is a rare, dramatic, and hard-to-hedge move.
So the banks went looking for counterparties willing to take that exact slice of risk - one-day extreme jump risk only. The deal worked like selling a stack of one-day options on a given name: if it drops 50% in a day, the seller covers the bank, and the seller agrees to keep providing that cover for two, three, or four months. Every one of these contracts added a steady bid for downside protection, which pushed volatility higher across a big sector.
The reversal
Once the pressure eased and the risk that forced the hedging faded, volatility ran out the door fast. Because these deals took heavy negotiation, they were signed as multi-month contracts, not day by day. So when the danger passed, the banks were left overhedged. They no longer needed the protection, and they had to sell it back and unwind the hedges. That selling has now pushed volatility lower than it probably should be. This has played out over the last two to two and a half weeks.
The result: volatility in the semiconductor sector looks a little light. Because chips are such a large slice of the S&P 500 and were the main thing holding index volatility up, S&P 500 volatility looks a little light too, all traced to this single source.
What comes next for chips
The boom-and-bust swing has cleaned itself out. The forced buying on the way up and forced selling on the way down from these leveraged players is largely gone. From here the sector should move more on real fundamental data than on forced trading. Boom-bust episodes are where volatility gets most dramatic and overcorrects the most, so volatility has likely fallen too far now from the forced removal of these hedges.
How to use it
The collapse in volatility is best treated as a tool for expressing a view, not a simple green light to pile back into chips. Since roughly mid-February, volatility had been very high. Anyone using options to bet on direction was often wrong even when right, because the options were just too expensive. Now prices sit at fair, maybe even cheap. That makes volatility usable again as a way to put on whatever view you hold.
