Doug Clark, managing director of liquidity research for ITG Canada, was recently interviewed on TabbFORUM in a post called “Dark Pools & Private Schools”. In his interview, Clark questions Canadian regulators for their recent regulation of dark pools. He compares the restrictions placed on dark pools to prohibiting parents from sending their kids to private school.
But his analogy is flawed.
In his analogy, the parents are the ones choosing to take their kids out of the public school sector, to place them into the private school sector. But in many cases the retail investor is NOT the one choosing to send their orders to a dark pool. In some cases, they don’t even have a choice. It is often the broker/dealer controlling the routing, and making that choice for them (usually based on the payments they receive from routing their orders to a specific OTC market maker or dark pool). Following Mr. Clark’s analogy, this would be equivalent to the education authority taking kids out of public schools and placing them into private schools without the parent having any say, and getting paid to do it!
Secondly, the reason the parent would move their child from a public school system to a private school system is to receive some sort of benefit – better food, better teacher to student ratio, better security, better programs, as Mr. Clark mentions in his interview. But what is the benefit that the customer is receiving by having their order routed away from the public exchange? The chance to receive sub-penny price improvement on their order? These fractions of a cent give little nominal value to the customer, but have a significant effect on deterring displayed liquidity, especially in thinly traded issues.
Consider the following example.
Assume stock XYZ has NBBO:
Bid Size Ask Size
25.50 500 25.70 500
A retail investor places a market order to buy 500 shares of the stock. The order is routed to an OTC market maker who gives sub-penny price improvement to the order, and executes the buy order at 25.6999.
Instead of paying $12,850 for the stock, the retail investor pays $12,849.95 for the stock, saving a whopping nickel.
But the participant who placed the 25.70 limit order on the exchange, the displayed liquidity provider, is left unfilled.
Imagine if you were that participant that should have received an execution, but did not because your order was stepped in front of at the moment it was about to be executed. This leaves you with an unquantifiable loss, the loss of the missed trading opportunity.
Now imagine a different scenario:
Assume the same NBBO:
Bid Size Ask Size
25.50 500 25.70 500
A retail investor places a market order to buy 5000 shares of the stock. The OTC market maker, knowing this order will impact the price significantly, does NOT step in front of the 25.70 order, and allows the market order to route to the displayed market, taking out the 25.70 seller, and trading all the way up to 26.00 to fill the balance of the order.
The 25.70 limit order trader is immediately down 30 cents on the trade.
The routing of retail order flow away from the displayed market has distorted the risk-return matrix of displayed market makers, and they are hesitant to make markets in smaller stocks as a result of the increased adverse selection risk.
In fact, the matrix has become so distorted that our firm, Bright Trading LLC, recommended to our traders that they stop providing liquidity in small and mid cap stocks entirely. The rationale is simple, if you get filled on your limit order, it is usually because you are on the wrong side of the trade, and the price often blows right through you.
Therefore, there is little point to displaying liquidity in these issues, as the return no longer justifies the risk. Our traders have learned this the hard way. The OTC market maker steps in front of your order when it is on the right side of the trade, and allows your order to be filled when it is on the wrong side of the trade.
We believe that the distortion of the risk-return matrix of displayed market makers due to internalization, is the main reason that we have liquidity problems in the small and mid cap space. It is apparent that the Canadian regulator (and now ASIC) have similar concerns.
To rectify the situation, the Canadian and Australian regulator have mandated that orders routed away from the public exchanges receive meaningful price improvement over the NBBO. That way at least there is some benefit to the end customer from the routing of their orders away from the public exchange. It should also help to deter OTC market makers from stepping in front of limit orders, as they now have to assume increased risk (by giving meaningful price improvement) when internalizing a retail market order.
But ASIC may go further. In their recent report on dark liquidity, not only are they mandating meaningful price improvement, but they are also proposing a trigger to implement a minimum dark order threshold, to apply to securities where there is evidence that dark liquidity has caused degradation in market quality.
These actions by the Canadian and Australian regulator should be applauded as opposed to questioned. Displayed limit orders are the building blocks of price discovery. It is imperative that we protect these orders. The new policy decisions of the Canadian and Australian regulator should help to improve the risk-return matrix of displayed market makers and encourage market participants to display limit orders more aggressively, which should improve market quality in these countries.
Every once in a while a key technical level becomes so relevant, that every time the stock moves away from that level, it gets pulled back towards it. I call these types of levels “magnets”.
Take for example the 55 level on LLY. For the past 8 trading sessions, LLY has found incredible support at this level. In fact, it has bottomed near the 55 level in each of the past 8 trading sessions. Every time the stock starts to pop, it gets pulled back by the force of the 55 magnet.
Technically, the magnet continues to hold for a number of possible reasons:
1) Traders have identified this area as support and continue to buy the stock in this area. As the stock rallies, those same short-term traders scalp out of their longs which doesn’t allow the stock to gain any significant upside momentum. Eventually the stock just collapses back.
2) Short-term traders have identified the relevance of the 55 level and continue to short the stock when it pops. Again the stock cannot gather any upside momentum to break the force of the magnet.
Regardless of the cause, if a trader can identify these magnet levels early on, there can be some significant opportunity to profit from these “Magnet Trades”, by simply fading the moves away from the level of significance.
Today a number of stocks had a mini flash crash exactly one second before the market opened. The affected issues: C, HPQ, T, WU, HES, WFC, KR, VTR, GS, S, RAI, WMB, CRM, IBM, LUK, & TAP
HPQ traded as low as $3.06 as you can see on the tape:
Today’s flash crash in these issues was caused by market fragmentation and inefficient order routing. These sudden market moves were caused when an order was routed “direct” to the Nasdaq (NSDQ), bypassing liquidity on the other exchanges. SEC rule 611, the order protection rule, is designed to prevent “trade-throughs”, which are trades being executed at prices inferior to the best-priced quotations. But this rule only protects the top of the order book, and the rule does not cover the pre-market trading session.
To better understand SEC rule 611, consider the following example:
Security ABC order book on Ask:
Size Price Exchange
300 25.00 BATS
300 25.02 BATS
300 25.03 NSDQ
NSDQ receives a market order to buy 600 shares of security ABC. The best available offer is on the BATS exchange at $25.00. NSDQ must route the first 300 shares to BATS because this order is at the top of the order book and is a “protected quotation” under SEC rule 611.
But here is where it gets interesting. Having satisfied the order protection rule by routing the first 300 shares to BATS, the remaining 300 shares is now routed back to NSDQ and executes against the NSDQ’s $25.03 quotation. The $25.02 offer on BATS remains unfilled. In other words, the second BATS order is traded through.
The participant sending the order actually gets an inferior price being filled at $25.03 on NSDQ when they could have been filled at $25.02 on BATS, but the $25.02 BATS quote (because it was not at the top of the book), is not a protected quotation and not covered by SEC rule 611.
In today’s example, the trades occurred during the pre-market session, so even the top of the book is not protected. With market liquidity being fragmented across multiple exchanges, order depth on one specific exchange can in some instances be pretty sparse, especially during the pre-market session. Sometimes directed orders are large enough that they can sweep through multiple levels on the exchange, causing the stock to have a sudden fall in price on that specific exchange.
This is what happened today on these 16 symbols. As you can see, it is very important to route your orders carefully in this fragmented market structure.
Note – Nanex has done an excellent job at analyzing the details of today’s flash crash event. You can view it here:
RIMM up 50% this month. HPQ up 10% since gapping down after the Autonomy debacle. BBY up 6.6% today.
What is powering the sudden surges in these perennial underperformers? Are these companies turning it around? Are these companies fundamentally better than they were a week ago?
Josh Brown at the Reformed Broker just answered this question in his blog post, “Here is an incontrovertible fact“, in which he discusses the real reason behind the sudden resurgence in these stocks. The answer is market sentiment.
Stocks move on short term supply and demand. When the majority of short-term market participants are on the same side of the trade, the price will almost always move in the opposite direction. The herd is always wrong. Too many participants have thrown these dogs out, and some real Johnny-come-latelies were recently shorting them. But sorry to break it to you Johnny, nothing goes straight to zero. There are always short covering squeezes, glimmers of hope, and market sentiment shifts. And the market sentiment has shifted for all of these issues. These dogs are barking.
You know what happens when one dog starts barking in the neighborhood, all the other dogs start barking as well. So don’t be surprised if some of these other dogs start gathering some short-term interest.
Advanced Micro Devices (AMD)
Alcoa Inc. (AA)
ITT Educational Services (ESI)
Apollo Group (APOL)
J.C. Penney (JCP)
The bottom line is that there may be some short-term upside in some of these issues, but don’t get married to any of them. Just because the dogs are barking, doesn’t mean they don’t all have fleas.
Many investment advisors advocate for the use of dividend reinvestment programs (DRIPS) in their client’s portfolios. Some brokerage accounts actually have the setting defaulted to automatically reinvest dividends into the underlying companies when a customer opens a new account. I believe this is a big mistake.
Take a look at the share prices of Best Buy (BBY) and Hewlett Packard (HPQ) today. They are making new 10 year lows. The business model of both of these companies is broken. However, both companies have had a history of paying a nice dividend. If an investor had been banking those dividends for the past decade, they may still be down in both positions, but they would have had a substantial amount of their initial investment recouped, just from the dividend checks.
The story is quite different if they participated in a dividend reinvestment program, as the majority of those dividend payments over the past decade would have been reinvested in the companies at much higher prices, eroding the value of all of those dividend payments.
The underlying problem with DRIPS is that very few businesses stand the test of time. Industries change, technology changes, consumer tastes change, and companies have to keep reinventing themselves if they are going to survive. In many cases this simply does not happen.
Just ask the shareholders of Eastman Kodak. It paid a healthy dividend for the past 50 years. If you had invested in the company 50 years ago, and had been banking those dividend checks, your initial investment in the company would have been worth a small fortune just from all the quarterly dividend payments you had received.
But if you had been participating in a DRIP, all of those dividend payments, and all of that cash would have been gone when Eastman Kodak filed for bankruptcy earlier this year. You would have lost it all.
DRIPS work great in companies that continue to grow and stand the test of time. But most companies do not. Placing all your bets that the companies in your investment portfolio will be there when you retire is simply a very scary bet to place. The best way to continue to build your portfolio is to bank those dividend checks and invest them in other stocks and other sectors. Those dividend checks are the natural way to diversify out of core holdings, because very few companies last forever.