Wall Street pros, if nothing else, are always in search of an edge. Whether it’s money managers looking to boost their returns for investors or brokers seeking to offer new or better services to their clients, the Street has always been a hotbed of innovation and competition.
Two firms-one sellside, one buyside-are staking a claim in the less-liquid names, in an effort to bring greater value to their clients. Jefferies & Co. has undergone a mini-transformation of its business model and is looking to commit more capital than it has in the past for small- and mid-cap stocks.
Meanwhile, Putnam Investments has reorganized its desk along the lines of liquidity characteristics, allowing traders to focus and specialize on the various market-cap segments.
At the same time, the environment for trading less-liquid names in the mid- and small-cap segments is evolving, with traders reporting that they are having more luck executing with algorithms than in the past. These algorithms stealthily access dark pools without tipping an investor’s hand, but they can also be more aggressive, depending on a trader’s strategy. Buyside traders contacted for this story say they are now electronically executing anywhere from 10 to 50 percent of their small-cap flow.
Venue Choices
That figure encompasses all electronic trading, including crossing, algos and traditional direct market access, whereby a trader posts orders and can submit hidden reserves. The latter remains an important part of small-cap trading. The same is true for independent crossing networks, like Liquidnet, Pipeline and ITG’s Posit, that can print size. With the increase in liquidity in numerous dark pools, traders say a symbiotic relationship between algorithms and dark pools has emerged, and that the lines distinguishing one from the other are beginning to blur. In the end, they say, it doesn’t matter where they trade, as long they can do so without market impact and avoid leakage of their intentions into the marketplace.
John Despotopulos, head trader at small-cap manager Lee Munder Investments, which has about $4 billion in equities, says he became convinced in the last year that he could effectively use algorithms for less-liquid stocks. “The intelligence behind algos has allowed them to better handle small-cap stocks, and they’ve gotten a lot smarter,” he says. About half of Despotopulos’ desk’s order flow is executed high touch, while about 10 percent is executed in dark pools, mostly via algorithms. So how does a trader decide where to go with an order? “It’s really a matter of knowing our stocks and who trades them,” he says.
Tereck Fares, director of equity trading at Chicago Equity Partners, says he expects the trading of small caps in dark pools via algorithms to continue to grow. “The liquidity in the pools is deeper today than it was a year ago, and it will only get better,” says Fares, whose firm manages $11 billion in equities.
Human Touch
Jefferies & Co. understands the buyside’s evolving outlook and can see dark pools and algorithms getting better at executing the less-liquid stocks. But the firm also believes there’s still a need for human traders to oversee the entire process.
And if volatility should return to the marketplace, traders are likely to become even more important for small- and mid-cap stocks. To be sure, the CBOE Volatility Index, a measure of market expectations of near-term volatility, has been unusually low since 2003, despite some shifts here and there.
But brokers and their institutional customers alike foresee the inevitable return of volatility, and some have taken steps to ready themselves.
Small- and mid-cap companies are generally defined as having market capitalizations between $300 million and $2 billion, and between $2 billion and $10 billion, respectively. Trades in some of these names have always been the hardest ones for which to find liquidity. Increased volatility would make that job that much harder.
Jefferies & Co., which does 75 percent of its business in the small- and mid-cap arena, has gone on a hiring spree in the last year, pulling over high-cost talent from competitors to ramp up its services in the realm of less-liquid stocks. A key criterion of those hires was whether the executives brought experience in committing capital-a significant change for Jefferies, which had been primarily an agency shop. “The new captains now running the sectors all have broad and deep capital-committing experience,” says Ross Stevens, co-head of equities at Jefferies. Stevens joined the firm in November 2005 from Banc of America Securities, where he was chief operating officer for equities and the head of electronic trading services.
Buyside Specialists
Putnam Investments, meanwhile, split its domestic trading desk into two groups about 18 months ago, charging one set of traders with focusing specifically on stocks with market caps under $4 billion. “We wanted our traders to develop greater expertise in those areas and work more specifically with portfolio managers and analysts involved in those types of funds,” says Richard Block, director of global equity trading at Putnam.
Block says the move succeeded in prompting traders to work more closely with portfolio managers to understand their investment objectives. It’s also enabled his crew to huddle and pool information about sellside firms-and to analyze which ones are most effective in specific areas. “We’re getting a more granular understanding of sellside expertise in different capitalization areas, down from the sales trader to the position trader, and which firms’ product offerings are strongest in different areas,” he says.
All the major brokerage firms are investing heavily in their electronic trading products. Stevens of Jefferies says his firm, in addition to hiring traders, has been analyzing trading patterns of illiquid stocks and developing algorithms to exploit them. He notes that unusual trading volume in a small-cap stock tends to persist, and that a “lead-lag” effect often takes place when a large-cap stock moves significantly. “Say a large cap jumps 2 percent. A small-cap stock in the same industry may also jump, but five, 10 or 20 minutes later,” Stevens says. He adds: “Algorithms can exploit that.”
In fact, Stevens says, Jefferies is building its strengths in the small- and mid-cap arena in large part for when volatility returns and buyside traders need that extra helping hand. Events earlier this year suggest that’s a good strategy. The 400-point drop in the Dow Jones Industrial Average on Feb. 27 resulted in the NYSE Hybrid market seeing its share in listed-stock volume jump to 70 percent, a fleeting reminder of how traders turn to the familiar in times of peril.
Being Prepared
Everything reverts to the norm, and when it does, we’ll use more brokers and more capital,” says an executive at one large money manager. Putnam is also counting on people power when volatility returns in force. Block highlights the relationship between higher volatility and higher transactions costs. In his view, by carefully monitoring the sellside and which brokers can best source liquidity at the most reasonable costs, Putnam will be ready when the period of reckoning arrives. “I think we’re positioned well for higher volatility,” Block says.
Best-Ex Stats
On the wholesaling side, Nick Ponzio, president of wholesaler Hill Thompson, is no stranger to volatility or illiquid stocks. He thinks electronic trading tools like algorithms and crossing networks are necessities for traders today. His trading desk uses them. However, Ponzio says, market makers are still important in the illiquid names and all the technology in the world can’t synthetically create liquidity. “A true market maker willing to commit capital where liquidity doesn’t exist still adds tremendous value,” he says.
Indeed, wholesaling is ultra-competitive. There are industrywide best-execution statistics-SEC Rule 605-that clients study closely. Joe Mecane, who heads UBS’s wholesaling effort, says brokerage clients, like any investor, want to see price and liquidity improvement, with minimal market impact in their executions. Bulge bracket firms like UBS have the ability to cross orders within their internal crossing engines, before moving on to the next step of committing capital, if necessary. “Capital is a very important component of achieving best execution,” he says.
But perspectives on the buyside are divided on whether an increasingly electronic and fragmented marketplace has made it easier or harder to find liquidity for small- and mid-cap stocks. The so-called smart money says algorithms will eventually connect all the dark pools. In addition, some dark pools themselves are in the process of connecting to each other.
Putnam’s Block says the increasing number of moving parts in the capital markets at least hasn’t made it more difficult to trade small- and mid-cap stocks. “We’re not finding it any more challenging today,” he says. However, he notes that Putnam’s new organizational structure may have mitigated some of the new complexities.
Being There
In the meantime, even with algorithms, there’s no guarantee that investors can be in every dark pool. Consequently, traders can miss stock if they’re not there. Of course, the easy answer is to give the order to a broker, which will ensure some presence. However, that comes at a higher commission than a self-directed electronic trade for an institution.
Will Sterling, head of institutional electronic trading at UBS, says he expects to see an increase in the usage of algorithms for small-cap stocks. For one, they’ve become better at accessing both displayed liquidity and undisplayed liquidity in dark pools. In addition, buyside traders are getting more comfortable with these strategies and have started to work with their brokers to customize algorithms that come close to how they like to trade. “It’s gotten to the point where algorithms can do a pretty good job in small caps,” Sterling says.
Not only have algos gotten smarter, but apparently, so has the industry. Small-cap algos are designed to be most effective when they react to liquidity, Sterling adds. They don’t work well using a benchmark that requires a rigid execution model like VWAP (value weighted average price) that attempts to trade at a consistent level regardless of price or available liquidity. A VWAP benchmark or an algorithm based on historical trading patterns work much better in the large-cap space than with small caps. To be fair, small-cap algorithms also analyze recent trading patterns, but the time frame is minutes or even seconds, compared with a large-cap algorithm, whose trading strategy could be based on trading patterns over the last 30 days. That works with large caps due to the greater frequency of trading and consistency of volume.
Balance Sheet
Committing capital is a tried-and-true way to garner flow from institutional clients, but the game is different in the small-cap space. First, brokers will typically “get an account started” on maybe 15,000 shares of a stock that trades between 150,000 and 200,000 shares a day. That usually means the account doesn’t get too much under its belt to begin with, says one buysider. And if there’s value in the idea, the account could lose alpha because it would have to step aside and wait until the broker gets out of its short position. Others, however, say they typically split prints with the broker.
The other problem with capital, particularly for small accounts, is that when a broker suffers a loss, the account has to both make up for the broker’s loss with additional commissions and still pay for its research commitments. Small accounts-if they can get capital in the first place-don’t have enough commissions to offset putting a broker in a hole.
There are varying opinions regarding the degree to which an account is supposed to make a broker whole after a loss. Some are of the opinion that “I didn’t ask for the capital-they offered it-and it’s their job to accurately judge the risk and price the merchandise appropriately.”
Cheryl Cargie, head trader at Ariel Capital Management in Chicago, says her best brokerage relationships would agree with that logic up to a point. Cargie says her shop will split stock with a broker after it makes the initial trade. “They don’t expect us to stand out of the way until they’re whole,” she adds.
Cargie says she’s getting more calls from brokers looking to offer their capital these days. “I’m starting to use it more only because I need to get my trades done,” she says, in reference to the fragmentation of the marketplace. “It appears to me, at least, that more firms are offering it, and they’re willing to use it more often. They want to get involved, to get something started,” Cargie says. She won’t always use it, she adds, but it’s good to know it’s there if she needs it.
Jeff Kaplan, a partner at $1.7 billion hedge fund Deerfield Partners, agrees that the sellside is at the ready to commit capital. The question, however, is whether it is in the account’s best interest to go that route. To get half of a 400,000-share order in an illiquid stock done “isn’t the worst thing in the world,” he says. But the last thing a buyside trader needs is to get capital for 50,000 shares of an illiquid stock and still have 800,000 shares left to sell or buy. The order will get shopped, and the information leakage will begin to impact the price.
“Being tied into the sellside with an illiquid stock that you own a large hunk of generally isn’t a good idea,” Kaplan says. Still, he adds, there is a partnership between the buyside and sellside. “They don’t call it a risk bid for nothing,” he says. “The last thing you want is to sell your broker 250,000 shares of a stock that doesn’t trade well, and then have bad news come out the next day-a surefire strategy to ruin your Street relationships.”
Greater Liquidity
Kaplan believes that the market is substantially more liquid today than it was before the industry shifted to penny pricing. A former position trader at Merrill Lynch, he says this is the case across the board, even for small caps. Granted, there won’t be the same size at the inside quote, but clients see much better menus with tighter pricing from brokers than they used to.
For example, say there’s a fairly liquid small-cap stock today bid at $20. As a seller, Kaplan might see a menu for a 250,000-share order of $19.90 for 25,000 shares; $19.80 for 50,000 shares; $19.70 for 100,000 shares and $19.50 for all 250,000 shares. All things being equal, in that same issue when stocks traded in eighths, Kaplan says he’d likely have been offered an inferior menu with much wider pricing. For example, the broker would have bid $19.875 for 25,000; $19.625 for 50,000 shares; $19.25 for 100,000 shares; or $18.75 for all 250,000 shares.”There’s no question the menu is substantially better today,” Kaplan says, pointing out the 75-cent difference per share in the last principal bid he would have gotten in each of the menus.
Still, there’s a difference between large- and small-cap stocks. “The difference between doing an OK job and a great job in a large-cap security,” according to UBS’ Sterling,”can be very difficult to measure. But the difference between an OK and a great job on a small-cap execution can be like night and day.” The reason, Sterling adds, is due to the volatility that small caps present and “the more sporadic liquidity.”
One buyside trader who trades both market caps says that if he can get 25 percent of his order done in the large caps “fairly tight” using capital, he will. But he never sees that type of support in the small-cap space. A large buyer might be advertising an indication to trade 25,000 shares, which isn’t an attractive option if the order is 500,000 shares. And he understands why. First, volatility in the large caps is nowhere near what it is in the small caps. “If you’ve got a half-million to buy, you can sometimes move the stock 5 percent on 20,000 shares,” he says. Second, brokers are much more familiar with the more-liquid names: They know how they trade and have a much easier time finding the other side with an outbound call from a sales trader. “In the small caps, it’s like two ships passing in the night,” the trader says. “It’s hard for them to find the other side, and they don’t want to commit capital to do the big pieces, anyway.”
Go Electronic
One buyside trader who trades for a $2 billion small-cap fund says he executes about half his volume electronically. He’ll use algorithms that scrape dark pools, crossing networks and DMA tools. “Dark pools are a great way to get an order started and remain anonymous,” the veteran trader says. “Sometimes I can get 25 percent of the order done before the stock even moves.” An immediate-or-cancel order is a favorite of traders to test waters to see what’s in the marketplace when starting an order.
The trader adds that he would probably do more electronic trading in small caps, but pension fund clients direct about 25 percent of their business to particular brokers for commission rebates. In addition, he also needs to pay brokers for his own firm’s research needs. Roughly a third of his electronic volume is done in the independent crossing networks like Liquidnet and others, while the balance is executed by algorithms accessing dark pools and DMA.
To a Level
When the algos aren’t having much luck finding contras in the dark pools and there’s no size in the crossing nets, the strategy becomes more traditional. “You’ve got to be aggressive and take it to a level,” he says, in reference to posting in the displayed marketplace. “Sometimes you can find pockets of liquidity.” The trader points out that there’s nothing wrong with moving the price up 17 cents in a $20 stock if the portfolio manager is projecting it to be a $30 stock in the near future. “I was around when moving it up an eighth or a quarter was considered a good job,” he adds, saying he’s amazed at how the industry has reacted to the change in tick size.
Still, that aggressiveness sometimes has to be tempered with patience, particularly if the trader is a buyer and the stock starts to run. “You need to get in as much as you can, but you also need to walk away if the pricing gets too silly for you,” he says. “When it comes down, you can build a nice position a lot of times.”
Along with algorithms, the block crossing systems are the first venue that traders check. Liquidnet, a buyside-only block trading system, averaged about 55 million shares a day in the second quarter. Pipeline Trading System is another favorite. Both dark books focus on small- and mid-cap stocks, with 71 percent of Liquidnet’s volume stemming from that category and 64 percent of Pipeline’s.
Match Rates
Over an average trading day, Pipeline says it finds block matches for mid-cap stocks 15 percent of the time and 10 percent of the time for small caps. Liquidnet claims matches occur in the system about 30 percent of the time. But clients, for whatever reason, step up to trade only about one out of four times, or 27 percent. That gives Liquidnet an execution rate of 8.1 percent, putting it in line with what many say is the industry average of 9 percent.
A broker-sponsored crossing engine like UBS’s has an average crossing rate of 15 percent. “Whenever we’re trading in a small-cap security, we’re trying to figure out how to gather as much liquidity as possible without signaling to the market, and a cross is one of the best ways to do that,” Sterling says.
But the world hasn’t gone totally electronic just yet. Count Lee Munder’s Despotopulos as not only a fan of crossing, but also of the expertise that regional or niche brokers offer. “They know their research, and they know where one or two sellers might be,” Despotopulos says. “It’s always nice to put up that big natural trade.”
