A quick search in the NYU Library’s Journals system for articles from The Economist that contained the words “dark pool” yielded the following:
- “Big headache or Big Bang?” 2007
- “The battle of the bourses,” 2008
- “Attack of the clones,” 2009
- “Rise of the machines,” 2009
- “Some like it not,” 2011
- “The fast and the furious,” 2012
- “Going broke in stocks,” 2013
- “Code blue,” 2013
- “The end of the street,” 2013
Over approximately seven years, dark pools evolved from a curiosity to a major upheaval in securities trading, and may actually be on the wane lately. Let’s take a minute to define the term. From “Rise of the machines” in 2009: “dark pools [are] electronic trading venues that conceal an order’s size and origin.” Here’s a longer definition:
Dark pools are places where anonymous buyers and sellers can trade directly with each other away from normal, “lit” exchanges. Because data are published only once trades are complete, institutional investors such as pension funds are meant to be able to take, or offload, large positions in quoted companies without alerting the wider market. (“Some like it not,” 2011)
Before dark pools, a buyer/seller with a large position in a stock would have the following options:
- just put the bid/ask on the stock market, taking any adverse price movement on the chin–this is the riskiest option because a large supply/demand shift can make a stock temporarily unsaleable (or permanently unsaleable in the worst case)
- have a broker or specialist do the bid/ask on the exchange but in small batches, say 100 shares at a time over a longer period, to avoid a sudden shift in supply/demand that sends the price rocketing in the wrong direction
- write an algorithm that does the bid/ask in small batches on the electronic markets (possibly on many electronic markets)
- call up colleagues that are known to have large positions in similar stocks and work out a private deal (a practice known as the “third market” or “block desk”)
Dark pools can be thought of as an automation of #4. Basically, these large traders form a consortium and agree to use custom software to search through their internal order books to match potential buyers with sellers. The stock bids/asks are never posted to the public exchanges, and the identities of the buyer and seller are not known to each other until the sale is reported, which happens after the sale is completed.
Several interesting things can be teased out of this list of options. Option #1 presented a critical difficulty: moving large positions without negatively affecting the price you get; that difficulty was ameliorated by option #2. But option #2 forced investors to rely on expensive brokers and specialists, whose intentions were never assured:
Institutional investors may complain about being forced into “dark pools” (off-exchange venues where they can deal anonymously) to avoid HFTs, but these pools existed before HFTs and were set up in part to avoid being scalped by brokers or floor traders. (“The fast and the furious,” 2012)
Along came option #3, once the rise of electronic trading made it easier to move large positions by clever automation of the role formerly occupied by the people on the trading floor. But electronic trading also made it easier for other algorithms to analyze activities like the series of small orders that indicated an attempt to move a large position:
The basic idea of HFT is to use clever algorithms and superfast computers to detect and exploit market movements. To avoid signalling their intentions to the market, institutional investors trade large orders in small blocks-often in lots of 100 to 500 shares-and within specified price ranges. High-frequency traders attempt to uncover how much an investor is willing to pay (or sell for) by sending out a stream of probing quotes that are swiftly cancelled until they elicit a response. The traders then buy or short the targeted stock ahead of the investor, offering it to them a fraction of a second later for a tiny profit. (“Rise of the machines,” 2009)
So dark pools evolved as a sort of peer to peer stock trading model, avoiding the risk and scrutiny of public exchanges. However, it wasn’t long before banks got in on the game:
Dark pools have proliferated over the past five years, winning volume from exchanges. Broker-dealers including Goldman Sachs have muscled in, setting up their own dark pools to capture transaction fees their clients would otherwise pay to exchanges. Pools owned by broker-dealers now dominate the sector, but as a result many of them have become much less attractive to large investors. (“Some like it not,” 2011)
And it wasn’t long before HFT got into the game:
Many dark pools that are owned by broker-dealers now welcome high-frequency traders (HFTs), who can use their speed to exploit price differences between exchanges and dark pools . An HFT might, for instance, lock in a high price by submitting a large sell order to a dark pool just as the exchange price of a share begins to fall. (“Some like it not,” 2011)
And now dark pools are showing various signs of declining popularity:
There are signs of a backlash from institutional investors. After several years of steep rises, trading volumes on dark pools have started to level off this year, according to the TABB Group, which tracks the industry. Some investors are only routing orders to pools that exclude HFTs. (“Some like it not,” 2011)
Alongside this is the decline of the public exchanges, and the increasingly marginal role of floor-based trading. In 2007, the New York Stock Exchange was riding high as a profitable, publicly traded company, buying its competitors, basking in the light of its deregulators. But fast forward to November 2013 to see the NYSE bought by an upstart electronic market competitor (ICE).
What does this brief history of dark pools tell us? Well:
- investors are not satisfied with limited liability–they want anonymity too
- large investors want to obscure the details of trades as much as possible
- HFT will infiltrate any electronic trading system that becomes available, particularly any that achieve high volume or high liquidity
- in the absence of real time public exchanges, banks will move to implement their own trading exchanges
From these trends, it seems clear to me that some form of crypto currency will succeed in this space because it has the two investor goals (anonymity and obscured transaction details) built-in.
It also seems clear that the plain old stock market used by regular people has become a second class place to trade. HFT (which can be thought of as meta-trading) and dark pools (analogous to Internet peering agreements) are the first class trading venues now. Regular people exist to pour money into the system by buying and selling, but also to pay the fund manager commissions, provide the HFT algorithms with liquidity rebates, and to pay HFT systems the “penny toll” that they extract from most trades they participate in. These new first class trading platforms will only be accessible by large players who have access to the best market data, research, and technology. Those large players will use these platforms to limit their risk, while the systemic risks are borne by the aggregated small investors on public exchanges/markets.
Finally, the typically cited primary goal of the stock market system (capital formation) has been pushed so far aside that one must wonder whether the model will persist. Stock prices have a diminishing connection to the performance of the company that issued the stock. Investors typically care only about making good bets–mostly short term bets, at that.
Let’s assume the stock market doesn’t exist. How will a company get capital without it? One answer is that they won’t, at least not in the millions and billions of dollars. Micro loans and other peer to peer lending systems seem like entirely more sustainable options for small business. Just as companies formed consortia to allow them to trade in dark pools, it is entirely possible that small businesses could form similar mutual lending arrangements within their geographic area, or even within their industry networks. For example, the owner of a local trucking company might recognize the benefit of a nearby warehousing company or a wholesaler, and might be willing to assist those other parts of the economic ecosystem with small loans.
As for regular investors, the question is how they’ll get the sort of investment returns they have come to expect (and plan their retirement around). Again, the answer might be that they won’t, at least not at the scale of the boom parts of our stock market cycle. Again, micro lending or geographically/economically local lending systems might be a viable option if they are able to pool enough capital, manage the overhead costs of a lending program, structure for small purposeful loans, and insure against default risks.
I’m not convinced that micro loans and peer to peer lending are the right path for a post stock market age, but they are an interesting start. My next dive will probably be into the micro loan world, starting with the work of Muhammad Yunus.