Buyside Warms Slowly to Options Algos

It’s a tough slog for brokers championing the electronic trading of options.

According to a panel of experts assembled at this year’s Options Industry Conference, the options industry is five to 10 years behind the cash equities business when it comes to servicing block orders with algorithms. While the big brokers are managing to cope with the rapidly changing market structure, they face roadblocks when it comes to broadening the appeal of electronic trading to their buyside customers.

"Our industry is in its infancy," Calin Ciordas, a director at Credit Suisse, told the OIC crowd. "Only a handful of firms have implemented time-based algorithms, and I don’t think the clients have expressed much interest."

Time-based algorithms are those that slice up large orders and feed them into the marketplace over the course of some time period. They account for about 70 percent of all trading in cash equities, but very little of options trading. That’s not to say algorithms aren’t used in the options industry at all. Market makers use algorithms to price their merchandize and brokers have devised one-off order-like algorithms. But the algo revolution that swept institutional block trading in equities has yet to arrive in the options industry. 

The catalyst for the revolution in cash equities, of course, was decimalization. Penny ticks caused spreads to collapse and size at the inside to diminish. That made trading orders in blocks impossible and ushered in the era of slice-and-dice algos.

With the four-year-old penny "pilot," the options industry is moving in the same direction but slowly. Spreads have shrunk, but liquidity is still available from market makers in size. Voice brokerage predominates.

There are other obstacles to institutional acceptance of algorithmic trading, according to the experts. For one thing, there is little use of depth-of-book data. Some brokers believe that if their customers could see all of the available liquidity on a screen, they would have more confidence in algorithms to access that liquidity.

"If we can get all of the feeds, establish a ticker plant and publish depth-of-book, the customer will have access to some indication of an average price," said Mike Rude, a vice president in Goldman Sachs’ electronic trading group. "They will know where they can get filled if they were to sweep the book. This is one of the things we want to look at."

Once depth-of-book becomes available, Rude added, brokers will be able to discern the amount of both displayed and hidden options liquidity. Brokers will be able to build algos to access that liquidity. And once the algos come on line, "you have the ability to measure. And once you have the measurements, people can start making slippage predictions. You generally get a lot more confidence about your ability to fill a large block order."

Gathering all that market data is no trivial task, however, the experts noted. Nor is it cheap. Most players still take a top-of-book feed from the Option Price Reporting Authority. It is less costly than taking in full depth-of-book feeds from all nine options exchanges, but can be slow and imparts less information. Still, given the tremendous number of quotes generated by options trading, a full depth-of-book feed from every exchange would generate a tsunami of data, the execs explained.

Also sidelining algo trading of options is the lack of an accepted benchmark, say the trading officials. Despite the emergence of GWAP, or gamma-weighted average price, the VWAP-equivalent for the options industry, uptake has been weak.

"We don’t really have a benchmark," Ciordas said. "GWAP is an interesting idea, but it has certain limitations. Actually, every type of benchmark people have tried to come up with has suffered from similar limitations."

Andrew Shultz, head of the strategic options business at Susquehanna International Group, agreed. "GWAP has gotten some press, but there is still a question of how you really measure impact," he said. "What is your vega impact? It’s not just the directional that makes those benchmarks much more complicated to come up with."

Despite all the hurdles to algo adoption, Ciordas, for one, is optimistic about the future. As penny trading makes more inroads into the options business, the need for algorithms will grow, he said. "With time, the options market will resemble the equity market in terms of volume that is algo-driven," he said. "People have to remember that developments in our market lag the equities market by five to 10 years. So it will take time for us to get there."