Stock-Trading Monopolies Weren't So Bad
The people who run Norway’s giant sovereign wealth fund have discovered something. It turns out that when stock trading is done in lots and lots and lots of different places, it can actually be hard for investors to get a good deal -- or at least to be able to tell if they’re getting a good deal. Report Bloomberg’s Jonas Bergmann and Saleha Mohsin:
“There’s a rent extraction from all these intermediaries,” Oeyvind Schanke, head of the fund’s asset strategies, said in a telephone interview on Friday. “We are trying to advocate that we need to bring some of this back to where we started by getting the participants to meet, creating a utility that’s there for the sake of transacting institutional-sized blocks.”
I have a crazy idea for what they could call this “utility”: how about a stock exchange?
Actually, that probably wouldn’t fly, given that “exchange” has come to have a very specific meaning. In a paper released last week on “Sourcing Liquidity in Fragmented Markets,” Schanke’s employer, Norges Bank Investment Management, defines exchanges as “transparent, ‘lit’ venues” where others can see your orders. Stock exchanges evolved to work in this way for all sorts of good reasons, but it has always been a problem for investors with big blocks of shares to buy or sell -- on an exchange, the price moves away from you as others in the market react to the knowledge that somebody is trying buy or sell a ton of shares.
As a result there’s a long tradition of “upstairs trading” that bypasses the exchange floor. Starting in the 1960s, as institutional investors such as pensions and mutual funds began to play the leading role in U.S. equity markets, brokers -- with Goldman Sachs leading the way -- built a business of block trading in which they took the risk of moving big blocks of stock. In 1986, Instinet automated this process with an after-hours dark auction that matched orders at the day’s closing price. Wrote Vladimir Markov and Tito Ingargiola in an informative 2013 paper:
This proved popular, as it allowed market participants to trade with little or no market impact when they did trade, and prevented information leakage when they didn’t find a match, as the orders were never exposed on a lit market.
Markov and Ingargiola both worked at Liquidnet (Markov still does), another such “block-focused dark pool” that was started in 2001. In the early 2000s, block trades accounted for a quarter of U.S. equity volume. With institutions’ share of equity ownership continuing to rise and new entrants such as Liquidnet devising clever new ways to match trades, you might have thought block trading would have continued to grow.
Instead, amid an explosion of regulator-encouraged and technology-enabled competition and innovation in equity markets, block trading collapsed. By 2012 less than 5 percent of the shares traded on U.S. equities markets were in blocks of 10,000 or more. Instead of trading in big blocks, institutions now often break their orders into much-smaller chunks that are then algorithmically routed by brokers to whichever exchange or other liquidity pool seems to be offering the best price.
Or something like that -- the several paragraphs the Norwegians devote to their algorithmic trading strategies in last week’s paper are not entirely comprehensible to a lay reader. The trading landscape they are trying to navigate appears to be the same high-frequency-trading-dominated world that Michael Lewis described in “The Flash Boys.” But while Lewis treats the rise of HFT as a morality play, from the perspective of a giant investor it appears to be just really confusing. The Norwegians think they’re getting a pretty good deal on their trades, and they’ve learned to avoid “HFT ping destinations” and other dodgy liquidity pools, but they would really like it if they could just go back to trading more of their shares in big blocks.
In fact, block trading has been making a bit of resurgence, both the old-fashioned broker-executed kind and the electronic version. Several new block-crossing pools have recently been launched, among them IEX, the startup at the heart of Lewis’s story.
But having more block-crossing dark pools isn’t necessarily better because, as the Norwegians write, “the probability of finding a match reduces geometrically with the number of venues.” In other words, as the number of block-crossing pools grows, the chances that you'll be able to find somebody willing to take those 20,000 shares off your hands in a given pool shrinks. You might think that competition would solve this by driving all the block trades toward one winning pool, but instead competition among block-crossing venues has led to “rapid changes in relative market share, driven by the fee structure as well as ancillary services offered by the venue.” One day one block-crossing venue is the best; the next day it's another one. It's too much for investors to keep track of. That’s why the Norwegians would like to see institutional investors get together and build or promote the building of a block-crossing utility. From the Bloomberg article:
“We’re in favor of trying to reduce the number of block crossing venues,” Schanke said. “One would probably be perfect.”
Now remember, this is how things worked back in olden times, when stock exchanges had something close to a monopoly on trading the shares listed on them. Regulatory and technological changes broke that monopoly. But it turns out that in stock trading, monopoly has its uses.
And yes, one of the new entrants is owned by Bloomberg.
This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.
To contact the author on this story:
Justin Fox at firstname.lastname@example.org
To contact the editor on this story:
James Greiff at email@example.com