Autocall dealers wary of Nikkei volatility surge

Dealers caught in danger zone as losses lurk on upside and downside spikes

Risk.net montage

Japanese stocks are edging into territory that is making exotic equity desks nervous. It’s the point at which a sudden market move – either up or down – threatens to drive up the price of volatility, right at a time when dealers would collectively need to buy more of it to rebalance their exposures.

Dealers say a rapid 5% rise or fall in the Nikkei 225 from 21,290, the benchmark’s level as of March 13, could lead to sizeable hedging losses for firms that sold autocallable bonds linked to the index. A fall of that magnitude or greater would be the worst-case scenario, according to one dealer.

“I think the Street could lose as much as double-digit millions in total here if there is a big fall now,” says Tomoyuki Sasai, head of Asia-Pacific solutions at Credit Suisse in Tokyo.

The danger zone reflects peculiar dynamics on the Nikkei 225, which is particularly susceptible to hedging activity linked to autocallable bond issuance. Dealers must dynamically hedge their vega profile stemming from the issuance – their sensitivity to a price change in the underlying as it moves between upside and downside barriers embedded in the structure.

On the downside, the market sits precariously close to the spectre of so-called peak vega – a shadowy inflection point in which normal autocall hedging activity breaks down. On the upside, a large and sudden jump in spot could cause products to knock out in unison, pulling a rug of volatility from under dealers’ feet. Either event would trigger a scramble for volatility – potentially against a backdrop of limited supply and rising prices.

In a late-February research note, analysts at Bank of America pinpointed at 20,000 the level of peak vega for Nikkei products sold to Japanese investors.

“Nikkei spot remains close to peak Uridashi vega and a sizeable move either up or down could make long-dated Nikkei vols bid,” stated Lars Naeckter, a Bank of America analyst and the note’s author.

Nikkei spot remains close to peak Uridashi vega and a sizeable move either up or down could make long-dated Nikkei vols bid

Lars Naeckter, Bank of America

Others believe the pain threshold may be as low as 16,000, though the exact level depends on where downside barriers are clustered in each dealer’s book.

On the upside, dealers note a glut of products would knock out between index levels of 22,000 and 23,000. An absence of institutional volatility selling on the Nikkei means autocallable issuance is the primary supply of volatility into the market. As products knock out, dealers are left with a short volatility position, which is typically replenished by new issuance. The problem arises if principal is not quickly reinvested.

“We see a decent amount of knockouts between 22,000 and 23,000, and as spot is going up, the duration of the products and vega depletion is real,” says an Asia equity derivatives analyst at a European bank.

Large dislocations in the structured products market are not unusual in Asia-Pacific. In 2015 and 2016, exotics desks lost money on hedges of autocalls sold in South Korea after the HSCEI index tumbled. More recently, French bank Natixis took a whopping $230 million hit from hedges on its Korean structured products book.

Getting caught out on the upside is more unusual, but exotics desks have been here before. A sharp 12% rally on the Nikkei in December 2012 was accompanied by a surge in volatility that left the Street with a combined $500 million of losses.

The Nikkei 225 is a common underlying for autocallable structures, known as Uridashi in the Japanese market, and often features in so-called worst-of baskets sold in abundance to retail investors in Japan and South Korea.

The structures typically have a three-year maturity and embed upside and downside barriers. If spot on the underlying index rises past the upside barrier, investors receive their principal and a generous coupon, while their principal is at risk if the lower barrier is crossed. The products leave dealers with a long volatility position stemming from an embedded down-and-in put option sold by product buyers to fund the chunky coupon.

As spot falls, dealers’ long vega positions linked to the issuance increases, forcing them to flatten exposures by selling put options. This happened in December when the Japanese benchmark plunged 15%. If spot falls below the peak vega threshold, vega exposure linked to the issuance falls away, turning issuers into buyers of volatility. This point was not reached in December, market participants say.

Since the start of 2019, the market has rallied. A jump in spot reduces dealers’ vega exposure relating to the issuance, forcing them to buy volatility. This can be met either through new issuance or buying back put options. If it is met via the latter, spot and volatility would move up in tandem.

“As we move away from peak vega, there is a high chance of a positive spot-vol correlation,” Bank of America’s Naeckter stated in the research note.

Panic on the upside

Once upside barriers are hit in a rally, dealers lose the volatility exposure from their autocalls as the products knock out, while their short volatility vanilla hedges remain. The rebalancing of exposures this situation precipitates puts upward pressure on long-tenor volatility. New issuance typically creates the required supply of volatility as autocall buyers are effectively selling put options to product issuers. As long as investors choose to reinvest the principal in new products when their legacy instruments knock out, dealers’ vega is replenished.

According to the exotics trading head at a European bank in Tokyo, this typically takes about two weeks, meaning a gradual rally would most likely not be too painful, as products knock out and reinvest in dribs and drabs. A sharp 5% jump could be problematic for exotics desks, however, knocking out a hefty chunk of products in unison and wiping out the market’s most significant supply of volatility – potentially for a fortnight.

“If it takes a couple of weeks, it is enough to be absorbed by new trades. But with one or two days it is difficult to be absorbed and it would have a meaningful impact on vol,” the trading head says.

Any dealer trying to cover its short positions against a dearth of supply faces a painful ride as vol is bid up due to the supply and demand imbalance, forcing dealers to buy at inflated prices and creating a vicious circle and costly rebalancing exercise.

“Back-end volatility is likely to be bid if we get a sustained rally as autocallable inventory would deplete and book duration would reduce,” says an exotics trader at a European bank in Hong Kong.

The real question is at which pace we see a market rally. If this is quite sudden, volatility would be well bid due to the supply and demand imbalance

Exotics trader at a European bank in Hong Kong

The trader agrees the speed of any move in the Nikkei is the crucial factor. If spot slowly edges higher, desks will be spared by new supply. A 5% move in one or two days, however, would see demand for volatility surge.

“The real question is at which pace we see a market rally,” he says. “If this is quite sudden, volatility would be well bid due to the supply and demand imbalance. However, if it is more progressive then clients would continue to trade new autocallable structures, and this vega supply would dampen the effect of the spot move.”

One such move has been seen on the Nikkei 225 in the past six months, when spot climbed by 4.81% between December 25–27, albeit off a sharp decline in which the index lost 15% of its value over the preceding three weeks.

Bharat Sachanandani, head of flow strategy and solutions for Asia-Pacific at Societe Generale, is convinced there will be a decent supply of new issuance in Japan over the coming weeks.

“Remember, all the comparative yield products have come off,” Sachanandani says. “US Treasuries hedged back into yen looks pretty poor; 10-year JGBs [Japanese government bonds] are back in negative territory. So where will the marginal flow of money going to go? We think the relative attractiveness of autocalls means issuance over the next few months is going to be pretty good.”

Dealers agree any damage incurred on Nikkei-linked autocall hedging is unlikely to be on the same scale of severity as earlier, high-profile events, even if the worst-case scenario was to play out.

“I think the size of the Natixis loss in Korea was unusual,” says Credit Suisse’s Sasai.