Monday, 18 April 2016

World closer to ending in tears: customers being paid to have a mortgage

Fire-fighting is dangerous
Negative interest rates have seemed a somewhat cute theoretical issue to many consumers. Negative interest rates? Perhaps just for the big end of town?

You may have grumbled that someone must be making out like a bandit as you continue to pay your mortgage.

No longer.

A funny / scary story in the WSJ [Negative Rates Around the World, Duxbury & Gauthier-Villars, April 14, 2016] relates the personal story of a Danish bank mortgage customer being paid by his bank to have a mortgage.

Here are some choice sections from the WSJ article:
AALBORG, Denmark— Hans Peter Christensen got some unusual news when he opened his most recent mortgage statement. His quarterly interest payment was negative 249 Danish kroner. 
Instead of paying interest on the loan he got a decade ago to buy a house in this northern Denmark city, his bank paid him the equivalent of $38 in interest for the quarter. As of Dec. 31, his mortgage rate, excluding fees, stood at negative 0.0562%.
“My parents said I should frame it, to prove to coming generations that this ever happened,” said Mr. Christensen, a 35-year-old financial consultant, about his bank statement."
And some of the consequences:
Prices of owner-occupied apartments have been rising. In Copenhagen, prices were 14.5% higher in the fourth quarter of 2015 than in the year-earlier period, compared with a 5.5% increase in 2014, according to the Association of Danish Mortgage Banks. 
In Stockholm, prices rose 17% in 2015, following a 10% increase in 2014, according to price-tracking company Svensk Maklarstatistik AB.
This isn't going to end pretty.


Thursday, 14 April 2016

Solarflare is suing Exablaze for patent infringement

Solarflare issued a press release announcing the court action against Exablaze for patent infringement. I can't help but not be not silent about this one, so let's meander through...

Criminal firms, such as Exablaze, deserve bad karma
Solarflare are the good guys in this battle. Exablaze is the quasi-criminal, nasty, patent-troll-like organisation that deserves your scorn.

I've read through but have not done a full analysis of the four patents subject to Solarflare's infringement allegations. Most of the patents have a priority date significantly before Exablaze was created. I know this as I instigated and managed the first FPGA NIC development at Zomojo, the firm that owns and span out Exablaze. So Solarflare have good priority at least.

The patents in question are linked here:
  1. US8116312-B2: Method and apparatus for multicast packet reception, Feb 8, 2006
    Claim of using filter to bypass kernel for delivery to user endpoint.
  2. US8645558-B2: Reception according to a data transfer protocol of data directed to any of a plurality of destination entities for data extraction, Jun 15, 2005
    Relates to usurping the OS by a user, or higher than OS level mechanism, for directing network data to an endpoint.
  3. US8817784-B2: Method and apparatus for multicast packet reception, Feb 8, 2006
    A continuation of US8116312-B2 regarding the establishment of packet filtering.
  4. US9258390-B2: Reducing network latency, Jul 29, 2011
    Using templates in a particular way to set up for transmitting protocols, filling those bits in, and subsequently transmitting the completed network message.
Exablaze's CTO Matt Chapman has commented on Solarflare tech in the past so his familiarity with it may indeed open Exablaze up to triple damages.

It is easy for me to pick sides as my I've expended considerable energy fighting against Exablaze / Zomojo. Greg Robinson, Matt Chapman, and their lawyers, such as Matt Critchley, deserve gaol time for the lying to the court, obstruction of justice, and other criminal acts. Matt Chapman lied to the court about a myriad of over 100 confidential claims. Chapman even tried to claim Sir Issac Newton's interpolation method was a valuable and confidential secret, proprietary only to his firm,
MR WOOD: I want to ask you a question about paragraph 98 of your MC1 document?CHAPMAN:---Yes.
HER HONOUR: Sorry, what? I missed that paragraph number. I apologise.MR WOOD: 98, if your Honour pleases.HER HONOUR: Thank you.
MR WOOD: Who is the Newton referred to in paragraph 98?
CHAPMAN---I don’t know how that’s relevant. That’s – it’s a well-known method for numerical approximate. It’s not Isaac Newton, no, I don’t think, but it may be. I don’t see how that’s relevant.
MR WOOD: I’m suggesting to you that it was Sir Isaac Newton’s method that was first published in 1685?
CHAPMAN: ---Okay.
Do you agree with that proposition?---That’s feasible. I don’t know.
It’s basic mathematics, isn’t it?---I would agree with that, yes.
Secret & confidential! Chapman and Robinson couldn't lie straight in bed.

In my action against them they had the benefit of being a friend of the judge and employing barristers who worked alongside the judge during the day. Terrible bias. It didn't help that the matter was considered so important that the eventual appeal was tacked onto a long sojourn by matey judges where one of the Full Bench jurists, Honourable (sic) Justice Gilmour from WA, turned up intoxicated in court (4th March 2015). There was never a chance.

Without a corrupt Australian court process, I expect Solarflare may fare better in the District Court of New Jersey. Also, being a jury trial, it may fairer. The full bench of the Federal Court of Australia did make a ruling that Greg Robinson was lying in their written appeal judgment. Exablaze has a history of lying to customers, staff, and courts. Perhaps Exablaze's lack of credility will help Solarflare in their just battle. A NJ court certainly should discount everything Exablaze, Chapman, and Robinson have to say. They are known to lie to courts.

Getting back to the tech, Solarflare make the #1 trading oriented network cards in the world in my mind. Simply put, if you want a high performance card with no fuss, use Solarflare. Having said that, I typically use Mellanox cards as the performance for me has been slightly better, with an improved cost profile, but with gritted teeth. From time to time, Mellanox have been a pain to get working properly, especially with in house apps. Solarflare, for me, is the “goto” card as its driver, OpenOnload(R), has been so much “cleaner” to work with. The Solarflare driver was very nice tech and I presume it still is.

It was interesting to read the Exablaze response to the court action. After Exablaze spent many years in court, lying to the court, perverting justice to misappropriate the intellectual property and patents of others, and cynically exploiting court processes, they had the temerity to include the following line in their PR response,
“While we respect any company's right to protect and defend its IP, we sincerely hope that Solarflare's actions in this instance are not a cynical attempt to exploit the legal process, rather than use innovation as a strategy to advance its competitive position.”
Yeah, right. What an ass-hat Exablaze's Greg Robinson truly is.

I hope karma has its day and Solarflare succeed in punishing the criminal firm Exablaze in court. I
Asshole firms, like Exablaze, sometimes get the karma they deserve
would also hope other firms, such as Exegy and Mellanox, who are likely to have intellectual property being violated by Exablaze, also take action. Indeed patent law says they need to take action as they otherwise surrender their rights. It would serve the community well to drive scum like Exablaze out of the industry.

I feel sympathy for Exablaze's customers and distributors who have been duped by Exablaze. Those clients may now be subject to having to return or give-up their Exablaze products. Such action may be hard to swallow but it may end up being a good thing, as there are certainly much better vendors to deal with.

Happy trading,


Saturday, 2 April 2016

A simple transport plan - make a few billion - save a few billion

I've long been a fan of high speed rail. No longer.

Let's meander through why autonomous-only roads may be better than rail, save billions of dollars, and make billions of dollars.

I've not much personal experience of high-speed rail, but I like it. I've only travelled on the impressive Maglev in Shanghai to airport train and a Shinkansen on the Osaka to Tokyo route. Only a slow 300 km/h in Shanghai (it will do over 400 km/h) and just a bit short of 300 km/h in Japan. The Shinkansen E5 series now will do around 320 km/h.

High speed railways are expensive to build and very expensive to maintain. In Japan thousands of workers are constantly doing track inspections and maintenance. Despite the high costs, the economic benefits of over 300M passengers per year in Japan piling onto bullet trains are obvious. That's a lot of time saved. It has been estimated that alone saves some ¥500 billion per year.
JR West machines from 2008
(click to enlarge, source)
Europe has some productive high speed rail. The US has long pined for it but has not been able to generate the will for even modest speed rail. Australia has long flirted with the idea of an eastern seaboard or Sydney-Melbourne link without any realistic momentum. It may have been lucky both countries have failed to launch as, just as with any technology, disruptive new ideas could change the game.

Want to make a few billion?

Perhaps a way to generate a few billion in profits, especially around geographically challenged cities such as San Franciso or Sydney, would be to have an interest in land that is difficult to have a sane commute to. Arrange the land and build high speed rail to grease the value proposition.

This is commonly referred to as "value capture infrastructure." For example, say you built a 500 km/h Maglev commuter to Willows CA from down-town San Francisco. That's about 225km (140 miles) by road or a bit under 200km as the crow flies. See the following map. Willows CA is inside the second biggest circle to the north of SF.
San Francisco with 50, 100, 200 & 400 km rings
(click to enlarge)
Imagine you bought enough land for 100,000 homes around Willows or a similarly distanced locale. Now build your MagLev. Say it has a top speed of 500km/h and you get a commute of under 30 minutes. That's pretty convenient. If your land is suddenly worth $100,000 more per block, or maybe $200,000 more, then that is a nice increase in value of $10B-20B. I'm not sure this is a great example, but you get the idea.

Earthquakes aside, it makes the economics of a 1000km/h Hyperloop to "nowhere in particular" pretty interesting, no?

Grow your city to 1M blocks of land and there is a lazy hundred billion in economic value to share between gov't, developers, and their customers.

A better plan

Elon Musk's Model 3 and other cars with autonomous capabilities, electric or not, provide a simpler solution. Imagine roads for only autonomous vehicles. No humans, so why not allow 200km/h travel?

Build a private road of 100 km in length. Own the property around the end of it. Profit.

Not enough autonomous capable cars? Have your autonomous car capsules ranked up and ready to roll just like a tiny train. Small packets of people cut down the latency of travel through quick start times.

A condition of travel would be autonomous only at the road's gateways. I'd certainly appreciate the freedom to work and travel without worrying about my safety or the inconvenience of discontinuous public transportation.
Tesla Model 3

It will take a while for cars to be able to legally self-drive on all public roads for good reason. Whilst autonomous cars will save lives and be a massive net gain, they will also cost the occasional life that a human may have avoided. In this Age of Perception, with its attendant disruption, higher level artificial reasoning will be slow to arrive. You would slow down when a beach ball had rolled across the road, even if it was well in front of you.  Your high level risk assessment would suggest  a child may be likely to follow. This is not the kind of logic a computer will easily carry out for the foreseeable future. Saving a thousand lives for the cost of one may be a good trade-off but an unnecessary loss is hard to swallow. This is an example of the kind of thinking that necessarily complicates and slows full autonomous transport adoption.

Governments could immensely improve the economic efficiency, safety, and liveability of their cities by allowing protected autonomous only lanes or roads. It is not just an acceleration of autonomous driving economic benefits. It is tangibly different. The main thing is with the microsecond attention span of the computer it will be safer to travel at very high speeds. You don't want human unpredictability interference in an autonomous environment. That is, even with autonomous driving, I don't want my car travelling at 200 or 300 km/h if someone may disrupt my life fatally by glancing at their text message on their phone. The only real and safe pathway to high speed driving is to take the human out of the picture.

Another benefit over high speed rail is that you will not need a few thousand workers inspecting short sections of road all the time. Maintenance costs would be much lower than high speed rail. We know how to build autobahns and other high speed roads. Autonomous cars can adapt to conditions and other vehicles through local sensing and communication.

This is similar, in some regards, to the Personal Rapid Transport idea that has been floating around for a long time. Well, since Donn Fitcher's work in 1953.

Autonomous commercial driving of trucks and buses will eventually displace the >11M trucks and buses on US roads too. Over 3M commercial drivers in the US will be redundant with more than 400M road hours of driving saved (2013 stats). That's quite some disruption with some sadness attached to the displaced until they find new, and hopefully better, work. Whilst road deaths in the US have dropped from over 50k to just over 30k, that many people dying on the road is just way too many. However, with not just >30k deaths, but more than 2.3M people being injured on US roads, there is a moral imperative to hasten autonomous driving.


Here is a map for the difficult terrain around my home in Sydney, Australia. Sydney's geography is such that it is one of the most expensive cities in the world to own property. Many suburbs, run-of-the-mill suburbs, have property prices over $AUD 1 million. Due to this property stress, many people commute for two or three hours a day just to get to work. Not much quality of life in that.
Sydney with 50km and 100km rings from the CBD.
(click to enlarge)
If you send a private autonomous road, or lane-way, toward Newcastle, Nowra, or Bathhurst, I'd imagine many people would be attracted by a short 30 minute commute in a 200km/h car from Katoomba, Lake Macquarie, Woolombi, Kiama, or Bowral with the savings in the price of housing. It would be faster than the existing trip for many people, like me, who live the burbs.

Wake up at 7:45am in Moss Vale. A quick shower and change. Jump in the car at 8AM. Get half an hour of work in on the autonomous commute into the city and get your early e-mails done. The car drops you off at the office. Your car then goes off and finds a park. After work, summon the car to the office. It picks you up and you relax watching the Nightly News on the way home and you're home by 5:45PM to spend some quality time with the family without the mortgage stress.

Perhaps it may be a Google-taxi or a Musk-pod? Maybe it is simply a government run electric car-pod for the main highway. It all sounds a big win to me.

Win win

Autonomous only roads. Good for governments. Good for developers. Good for car manufacturers. Good for freight. Good for consumers. It's good economics. Please get on with it.


Saturday, 26 March 2016

ITG continues to deceive customers

Despite ITG's wows of recent times, the firm continues to take advantage of customer gullibility.

ITG engages in proprietary trading. ITG markets itself as an agency only business.

After their little incursion with the SEC that cost around $20M for some paltry trading returns, ITG has not been called to account for the $70-$100M in proprietary trading profits they have made over the years despite their agency-only claims.

In fact, the ITG 10-Q filings with the SEC were a little thin on mentions of proprietary trading after their fine. It should come as no surprise ITG were a little sheepish about further prop trading disclosures. Nevertheless, they are now bolder. Late February ITG filed their annual 10K. It reiterated their ongoing proprietary trading function in the inter-listed arb (US listed v CAD listed stocks) on page 35:
"Other revenues increased due to higher principal trading gains on inter-listed arbitrage trading."
Canadian Operations
Year Ended
December 31,
$ in thousands
20152014Change% Change
Commissions and fees
Total revenues
Compensation and employee benefits
Transaction processing
Other expenses
Total expenses
Income before income tax expense

Canadian "Other" revenue, once again, was higher than "Other" revenue for the rest of the planet.

$8M each year for the last couple of years is nothing to be sneezed at. How much of that trade is facilitated by or within their own pool facilities? A former ITG Asia-pac CEO told me that some of it is due to ITG pool trading. This CEO was not a trustworthy source about many things but maybe he was right about the incursions into the client pool?

It's worth remembering some of specifics of the SEC findings that resulted in that fine:
6. While Project Omega was engaging in proprietary trading, including with ITG’s own customers, ITG was simultaneously promoting itself, and POSIT, as an independent “agency-only” broker that did not have conflicts of interest with its customers and that protected the confidentiality of its customers’ trade information.
8. ITG Inc. and AlterNet violated Sections 17(a)(2) and 17(a)(3) of the Securities Act by engaging in a course of business that operated as a fraud and by failing to disclose to ITG customers and POSIT subscribers, among other things, that: (i) ITG was operating a proprietary trading desk while at the same time promoting its brokerage services and POSIT by describing ITG as an independent “agency-only” broker;
13. ITG has historically operated and marketed itself, and has a reputation as, an independent “agency-only” brokerage firm. This designation was meant to convey that the firm did not engage in proprietary trading for its own account.
Even the SEC findings make it seem like ITG only engaged in proprietary trading for a short time:
ITG Launched a Proprietary Trading Desk in Early 2010. 
22. During the period of late 2009 to early 2010, ITG explored initiatives to increase
diversification and revenues for the firm, including launching a proprietary trading operation that would engage in algorithmic high frequency trading. Thereafter, on the recommendation of senior management, Group’s Board of Directors approved a proprietary trading desk that was limited in scope to inform whether ITG should launch a fully-scaled and disclosed proprietary trading operation.
That is clearly not the case. Let's have a look to the Canadian proprietary trading that traded US listed and Canadian listed stocks for well over a decade. Most of the revenue in Canadian "Other" is prop trading from inter-listed arbitrage. Here is the list I scraped from the SEC 10-K and 10-Q filings:

$ in thousands
Y2015 8,295
Y2014 8,066
Y2013 8,193
Y2012 6,020
Y2011 6,282
Y2010 5,572
Y2009 8,450
Y2008 16,512
Y2007 11,042
Y2006 8,800
Y2005 increased from what?
Y2004 ?
Y2003 ?
Y2002 ?
Y2001 ?
Y2000 ?

Total 87,232

Whilst ITG deceived customers with the "Agency only" lie in their marketing, the actual trading has been hidden in plain sight with the following snippets from their SEC filings:

Y2015 Other revenues increased due to higher principal trading gains on inter-listed arbitrage trading.
Y2010 Revenues from principal trading (included in other revenues) were lower in 2010 as the expanded presence of professional trading firms has significantly reduced the spread to be earned and thus limited principal trading opportunities available.
Y2009 Revenues from principal trading (included in other revenues) were down in 2009 compared to 2008. Aside from an unfavorable foreign exchange impact of $0.6 million, the reduction was attributable to several factors including lower market volatility and the expanding presence of high frequency participants (which has narrowed spreads and therefore reduced profitable trading opportunities).
Y2008 Interlisted arbitrage trading revenues improved to $15.9 million compared with $10.9 million in the prior year, benefiting from higher market volatility.
Y2007 Other revenues included $10.9 million from our interlisted arbitrage activities versus $8.8 million in 2006
Y2002 (e) income/loss from positions taken by ITG Canada as customer facilitations (a customary practice in the Canadian marketplace) as well as income from same day Canadian interlisted arbitrage trading.

Some laughed at how inept ITG seemed at proprietary trading with Project Omega not reporting significant profits. North of $70M is not trivial for this lot of proprietary trading.

Disgorgement and a large fine would be painful for ITG.

The many good staff at ITG deserve better treatment by management so they don't continue to deceive their customers and betray their trust.

SEC & IIROC, over to you.

ITG POSIT Finra ATS Tier 1 Statistics
1TG at #14 and approximately half of pre-scandal levels
(click to enlarge)

ATS data is provided via and is copyrighted by FINRA 2016

Wednesday, 23 March 2016

Most people in Australian tech firms at risk of 10 yrs incarceration from April 2nd, 2016

If you work in a bank; if you export a cloud service from Australia; if you are Australian and work overseas and use encryption; if you directly or indirectly use a processor >40MHz such as any smart phone... you may be headed to gaol for up to 10 years from April 2nd 2016.

Do not pass go, do not collect any investment returns, unless you have Australian Government approval for your situation, service, or product.
Most Australian and foreign businesses may have significant
additional legal risks soon. Whether you're a tech start-up or a
bank, be careful out there.

Once you stop laughing and take in the seriousness of this stupidity, if you're an Aussie or otherwise interested party, sign the petition here against this abomination.

From New defence trade controls threaten academic freedom and the economy
The Defence Trade Controls Act (DTCA) goes into effect on April 2, 2016, and it applies to all Australians, including those now living overseas.
The DTCA brings in a new regime of Department of Defence (DoD) oversight for both military “goods”, meaning new scientific ideas and means of application, and “dual-use goods”, which are innovations that may have some military use. 
The DTCA introduces a permit regime for any “intangible supply” (especially electronic communication) of new ideas in DSGL areas. Researchers and innovators who communicate any new idea overseas without permission face ten years in prison and A$400,000 fines. 
In other words, if you deal in new ideas in any of these areas, and you do not apply for a DoD permit, you are putting yourself at serious legal risk. 
The DSGL is clearly difficult to maintain. For example, it refers to integrated circuits running at 40 MHz or above, which were state of the art around 25 years ago. Recently Daniel Mathews pointed out that the DSGL controls encryption using only 512 bits, also long obsolete.
From Paranoid defence controls could criminalise teaching encryption
The bar is currently set low. For instance, software engineers debate whether they should use 2,048 or 4,096 bits for the RSA algorithm. But the DSGL classifies anything over 512 bits as dual-use. In reality, the only cryptography not covered by the DSGL is cryptography so weak that it would be imprudent to use. 
Moreover, the DSGL doesn’t just cover encryption software: it also covers systems, electronics and equipment used to implement, develop, produce or test it. 
In short, the DSGL casts an extremely wide net, potentially catching open source privacy software, information security research and education, and the entire computer security industry in its snare. 
Most ridiculous, though, are some badly flawed technicalities. As I have argued before, the specifications are so imprecise that they potentially include a little algorithm you learned at primary school called division. If so, then division has become a potential weapon, and your calculator (or smartphone, computer, or any electronic device) is a potential delivery system for it.
The DTCA is in essence an attempt to clamp down on research, theories, essentially ideas, that could have possible military applications. The idea itself is nothing new; Australia’s latest attempt is merely an incredibly potent example of how inept governments can be at dealing with these concerns.

With Australia’s new Prime Minister, Malcolm Turnbull’s “innovation agenda” which aims to effect a major structural change in Australia’s economy by fostering high tech industries, and attracting foreign investment into the high tech sector one must say Australia’s government seems to be at cross-purposes with itself. DTCA will kill these plans and it seems unclear if anyone has actually explained this to the new Prime Minister.

Saturday, 30 January 2016

Jon Flanagan - LFC 2016

“It was always going to be mentally tough. I got told I wasn’t going to be out for as long as I was and to keep going for the length of time I have been out it is hard to get your head around." 
“You always have your moments, but you just have to stay positive. I think I am a strong character. You cannot go around with your head down feeling sorry for yourself, you have to pick yourself up and stay positive really. I always knew that I would overcome it.”
Scouse inspiration on the value of hard work from LFC's 23 year old Jon Flanagan January 2016 after 619 days of injury and recovery[source].

Friday, 15 January 2016

One for the road: Failure-to-deliver, Goldmans & Reg SHO

I've long worried about the craziness of the US market with respect to fails-to-deliver, so I couldn't resist a new piece of SEC news where Goldman Sachs has been fined $15M.

As a trader when you trade via a broker you normally have an ETB and Restricted list to comply with.  For whatever reason, you can't trade stock on the Restricted list. An ETB is the broker's Easy-To-Borrow list of stocks that are OK to short as there is plenty of liquidity for loans if you're short at close of business. Often the ETB has quantities that represent the start of the day position but you can never really be sure at the close. It's not usually much of a concern to an HFT as your absolute position sizes shouldn't be too high as you churn the pennies out. You just stick to the stocks on the ETB list.

When stock or money is not delivered on the T+3 settlement there is a "Failure-To-Deliver".

So, Goldman Sachs have just been fined $15M by the SEC for violations related to this. Here is the SEC document. It makes for a fun read as it is easy to understand how the automated mundane of unquestioning workflows, and pressing the "F3" key, can arrive at such a point. The F3 key here turned into the FU2 key. It's the kind of boring detail that gets stuck in some business processing re-engineered workflow that gets ignored as people do interesting things until it rears its head and bites you in the ass to the tune of $15M. That's some bite.

I've long worried somewhat pointlessly about this.

The SEC publish fails-to-deliver twice a month. Let's look at the total share volume of fails for the last couple of full months:

October 2015 3,785,306,033
November 2015 3,722,780,135
update: Dec 2015 4,273,095,148

That's a lot of shares that don't settle on T+3.

Infact, it is more shares than a modest broker, who may be getting smaller, like ITG, may trade for an entire month. Here are the ITG numbers for November 2015:

ITG U.S. Trading Activity

# of

Total U.S.

Average U.S.
Daily Volume



Alert Daily

Trade Size

Avg. Trade
Size Ex-

November 2015


















Reg SHO is the weird piece of regulation that gives T+4, T+6, T+13 cascading forgiveness plus there is a DTC way of loaning and covering some of the madness. Unfortunately, some in the industry treat settlement as entirely optional. Sometimes it is because you'd prefer to cover that short in a day's time at a better price. Sometimes it is just because someone forgot something. Then there is just the poor practice of workflow rot that seeps through the cracks into a place like Goldman Sachs when everyone is trying to do things not so boring.

There are legitimate reasons why a failure-to-deliver should not be a huge deal but it should also not be settled for as a common settlement settling. Perhaps charging $10 per share or 10% would wipe out the practice smartly but that seems extreme. My prediction for 2016 is that this will largely be ignored and swept back under the carpet because it is so not so interesting except to ageing fintech geeks like me.

Happy trading,


Thursday, 14 January 2016

Happy new year and bye

It's 2016 and I've been sitting on the bench for a long time. Time to get back to work and feed the family. This somewhat unprofessional blog, and other media, will no doubt go as a consequence.

In the meanwhile, if there are any nice projects you may think would be mutually beneficial, reach out.

Happy trading,


CME to Shanghai - Uncle Kim?

Transit negotiations for the lowest latency path between CME and Shanghai would be interesting.

You'd better hire Dennis Rodman as a lobbyist.

Direct Path: Chicago to Shanghai via North Korea (Source)

Uncle Kim and Dennis hanging out: Source

Saturday, 2 January 2016

IEX - InvestorSexChange - the quest for validation & revenue

OK. Let's cut to the chase.

Q: Should IEX be allowed to become an official stock exchange?
Me: Yes

Q: Should IEX have protected quote status?
Me: No.

Huh? Those things are in conflict. How can it be so?

Firstly, I have nothing much new to say that hasn't been said already. If you're expecting some insight here, you should probably stop reading now and find a better way to start 2016. Here are some of the more educated opinions from people clearly smarter than me:
I'm not sure anyone cares for my thoughts on the matter. I've stayed quite quiet but a few people have asked politely for comment, so here are my meanderings.

Firstly, the IEX has been chugging along quite well. As you can see in the chart below, they are currently the third biggest ATS in tier one stocks by volume as at December 7, 2015 (coincidently my start to the 25th year of marriage, poor woman), according to the latest published statistics from Finra. They remain a small player in the overall market being reported to be around 1-2% of overall volume. Funnily enough IEX's biggest traders each day are usually HFTs as confessed by Citadel.
IEX - third largest ATS in tier one stocks (click to enlarge)
ATS data is provided via and is copyrighted by FINRA 2016


The positioning from those for and against IEX's exchange application has been pretty strong and vocal. Much of the debate, especially the Joanne Doe letters from many investors to the SEC as comment letters, has been rather silly and pointless, though often funny. Here is a link to the SEC letters.

The spoof comedy letter from a fictional town in Oregon with someone pretending to be a kid wanting IEX's application to be approved is perhaps the best example of facetious interest in the matter:
Subject: File No. 10-222
From: Danny Mulson
Affiliation: 8th Grade Student
December 15, 2015
Dear SEC,

I am a future stock investor, currently in the 8th grade at Aberdeen Middle School in Wetlawn Oregon. I whole heartily approve of the IEX plan to slow down trading.

Things move too fast in this world and we need to slow it down in every way we can. Take my school for instance. When we stand in line to pay for our lunch, we have two cashiers to pick from. There is Mr. Fields, who was recently fired from a data entry job. I heard he made some creepy comments to one of his female co-workers. But that isn't important. Mr. Fields can work that cash register like nobodies business. Even when twice as long, most kids will get in Mr. Fields line. Me, on the other hand. I prefer to get in Josephine's line. She used to work at the DMV before failing a drug test. I always get in her line. Sure, it takes me much longer, but she will always give me a compliment and ask me about my day.

In conclusion, IEX should be rewarded for slowing everything down and moving things backward. Cause backwards is awesome

Thank you for your time.
I'm pretty sure "Danny" is not referring to the SEC's Mr Fields and Josephine isn't a pun on Mary Jo White. Surely not. A follow up letter from the same fictional school in Wetland demonstrates why the SEC should have a little more rigour in the identification of bona fide credentials for letter acceptance.

Subject: File No. 10-222
From: Emma Hibernia
Affiliation: Aberdeen Middle School
December 23, 2015
Dear SEC:
I saw that letter Danny Mulson wrote to you. The Whole School Saw It. You need to know that Danny Mulson NEVER tells you everything he just tells you what HE wants you to know. 
OK Mr. Fields is fast. BUT Mr. Fields also takes 10 cents from everybody he checks out and when you complain about it he just laughs and says it's part of his job and everybody gets paid for their job so he should too. I thought he already got paid? But he said it wasn't enough. And suppose you want TWO milks well the first milk will cost you one price and then he charges MORE for the second one and you say they were all there to start with and the price tag was the same for ALL of them and he just says things change. WHAT CHANGED?? 
And Mr. Fields is creepy too just like Danny said but he didn't tell you everything at all. Mr. Fields got the cafeteria manager to make the GIRLS go down one ramp to the registers and the BOYS go down another ramp and the girls ramp is LOWER so instead of looking at us at eye level he gets to look down at us. Do you get what I mean?? So we complained to the cafeteria managers and they said it makes everything BETTER? And even though I begged her NOT TO my mother complained to the school board and they said they would put it on the agenda for like some meeting a gillion years from now after they talked about parking and everybody is already DEAD. 
Very truly yours,
Then you see part of the problem with a well meaning letter from a Professor of Physics, if a true letter that is,
Subject: File No. 10-222
From: John B Rundle
Affiliation: Professor of Physics, University of California

December 31, 2015

I strongly urge that IEX be granted the status of a national exchange. Please do not let the opposition of an entrenched few competitors roadblock their application. All investors want and need increased transparency in trading of securities.
Finance and market structure seems a field far removed from Professor Rundle's field of earthquakes. Michael Lewis wrote a largely fictional work in Flash Boys where the old mates of Lewis from one of his prior fawning books, "The New New Thing" (a fun read), invested in Spread Networks and IEX. The stupidity espoused as fact in that book, and 60 minutes subsequent "insights", has stirred the pot with many readers believing the crap about Reg NMS being broken because Michael Lewis told them so. Perhaps Professor Rundle and many other letter writers fall into this camp of mistaken fan girls and boys. If the Lewis story was correct their outrage would be well placed, but the Lewis story is an improperly researched fraud perhaps biased by his prior friendships from parties related to "The New New Thing." As an ageing HFT guy I certainly have my own prejudices.

Professor Rundle's letter is amusing as much of the benefit in market structure from IEX, yes there is benefit, comes from the DPEG order type. IEX's DPEG is complex, hidden or dark, and not necessarily an order a retail customer would easily understand nor lodge via their broker. In December 2014 IEX claimed its use was increasing and was then up to 11% of volume. This kind of order is beneficial to many users, not so great for some others, but parasitic, like most dark orders, with respect to market structure. I don't think Professor Rundle understands this opaqueness based on his short letter referring to transparency.

I wrote disparagingly about the IEX DPEG a while ago saying that IEX should not be allowed to become an exchange with such a proprietary order type that had such unknown workings. It was a good example of IEX espousing virtues of simplicity, transparency, and openness and not walking its own talk. I'm glad to say IEX largely fixed those issues of "magic" workings with a more detailed explanation of the DPEG in their exchange application, though they failed to properly disclose timing related issues. Ironically, this order type is exactly the kind of complex order Michael Lewis and Brad Katsuyama were both rallying against in Flash Boys. One woman's terrorist is another woman's freedom fighter.

IEX benefits

The DPEG is a pretty good order for HFT types. In a simplistic sense, it may allow you to sit on the bid or offer and be protected from some adverse selection with DPEG's last-look like facility. This is due to the fact that IEX uses current market data without any delay for crumbling quote protection versus stale magic shoe boxed data otherwise. DPEG is parasitic to market structure as it is only valid in a market environment that provides an appropriately efficient context.

Apart from that benefit, I can't really see any benefit from IEX apart from the fact that they have good ethical people running the exchange, like many other exchanges, with a particular customer segment, institutional investors and HFTs, they suit best. HFTs and IEX? Yep, remember Virtu was one of their larger early customers and in Citadel's comment letter they noted that on some trading days Citadel themselves were the biggest trader on IEX. 

Fundamentally I believe IEX has a very poor business model as it sails against the forces of efficient price discovery and risk transfer. Historically inefficiency in an enterprise makes for a shortened life span.  Once you also figure in their more expensive than normal transaction fees, long term survival seems unlikely. They have great marketing and mind share, as you can see from the letters submitted to the SEC, thanks largely to the effect of Mr Lewis and Flash Boys. At the end of the day, failure or adjustment of course seems the only likely outcome for IEX but that may be years away. They have good momentum.

The mistake that became IEX

In Flash Boys, Lewis describes Brad Katsuyama as a broker that developed a new insight into the market that led to the tool Thor at RBC. Thor could hit every bid or offer in the market by timing the order releases so the child orders would hit every exchange concurrently. So much for phantom liquidity. Turns out it was real. Basically Brad was being paid $2 million a year to execute badly for his clients as many others in the market already knew this somewhat obvious fact. So Brad mistakenly thought he was on to something and decided if he could slow down every order at a new exchange then he could prevent his customers being picked off. So, off he went and created IEX on this dumb premise. It was dumb premise as slowing down the orders just makes you an inefficient and slow exchange. You're still susceptible to faster traders picking off orders as it remains a race to the order queues as I pointed out previously and Citadel also noted in their SEC comment letter(s).

The eventual development of the DPEG order added a benefit due to the differential nature of the use of use of market data. Using current market data when others see delayed market data is the same as seeing into the future. So DPEG is useful to some, but it has to be understood to be parasitic and thus unworthy of market defining use in the large. That is, there is nothing wrong with parasitic if it is useful and doesn't destroy the host where the host is the NMS in this case.

A truly fair and efficient exchange

The best exchange for retail and institutional investors would look somewhat different to IEX but largely solve the same problems that Brad set out to solve. It would have the following features:
  • the lowest match and report latency (efficient)
  • length matched cabling in a co-location facility (fair)
  • no feed back on order lines, e.g. one way UDP in (no canaries)
  • common market data for all that includes coded order feedback
  • low latency microwave market data piped in, including CME (assist fairness for all)
  • simple order types, add, delete, modify for limit/ioc and that's about it (think Nasdaq UFO/Ouch)
  • carefully calibrated, low, and simple transaction pricing with no volume level breaks 
Plus a few other features. This would make for a seemingly simple exchange that would have the winds of efficiency in price discovery and risk transfer at its back providing a natural liquidity sink and business momentum. Ironically HFTs would not like such an environment as it would be very difficult to get any kind of competitive edge on their brethren. However, many HFTs also would like a simple exchange where there was no obvious disadvantage to blind side them. That is a constant HFT theme. HFT's like to find an edge on their competition, but prefer low latency simple environments without disadvantage where it is ironically hard to find any edge.

If you look at the above list and count the number of those features IEX has, you quickly get to zero. IEX could do better. There is still space for simple innovation in the exchange space at least.

OK, OK, get to the point, should IEX be an exchange?

Yes, I think IEX should be allowed to be an exchange based on my principle that it offers a bit of innovation with the last-look DPEG, it is acting honourably, and it is no worse than other platforms.

IEX's "look at our pretty colour" marketing does not appeal to my inner engineer that wishes for proper purpose. However, just because the IEX business model is a bit daft and its presence is suggestive of petty pretty pointlessness doesn't mean it shouldn't be allowed to be an exchange. IEX should be allowed to be stupid and fail. Stupidity is not a crime.

I certainly don't think IEX should be allowed to be an exchange with the non-delayed order routing by its broker dealer. We don't need an escalating war of delayed routing between exchanges screwing up the NMS. However, this could be dealt with by the SEC in a conditional approval. Similarly I think IEX's original application was flawed in terms of disclosure around workflows with respect to timings and delays but this was largely corrected, when forced by complaint, in further submissions. The poor form of the original application in this regard could result in the SEC denying and asking for a fuller reapplication, perhaps with a shortened process if that is allowed, but I don't think that is necessary given the widespread debate and current information. Minor corrective submissions or comment should be enough.

The purposeful delays in IEX's system should also make it that so they don't get protected quote status for their prices within the NMS. Quotes should be tradeable. A solution, such as Thor, should find real executable volume to the best of an exchange's ability and not a significantly and purposefully delayed rear view mirror of the past. If you want to read about protected quotes in Reg NMS Sec 242 Rule 611 and friends, please do. Simply put, exchanges' quotes are protected in that every broker must seek the best price amongst exchanges for their clients. You have to go to the exchange with the best price even if you think it is a tinpot exchange and not necessarily in the best interest of your client. The delays inherent in IEX should not force that upon the required parties.

That is kind of interesting as I don't think it even makes sense to be an exchange if your quotes are not protected quotes. It just makes you an ATS. Also I don't think the SEC even has the latitude to make such a ruling within the context of Reg NMS. This would mean that the SEC can't give IEX exchange status. However I think this reflects poorly on Reg NMS rather than IEX. I think IEX should be allowed to be an exchange with the basic conditions I've cited. Likewise participation in the consolidated tape revenue should be seen as problematic for IEX due to their delayed quotes.

Other arguments

Some people think the 350 microsecond delay IEX put on their system through a common POP in NJ should be enough to kick them from consideration. Reg NMS doesn't support purposeful delays. IEX replied that exchanges already delay in their co-location centres by having standard cable lengths and thus delaying some participants for fairness. Both sides seem a bit wrong there. The 350 microsecond delay is not comparable to the around 300 nanosecond delay implied by extra cable length. There is a 1000 times difference. Hardly comparable.

Even so, 350 microseconds of delay should not be a big deal in the current Reg NMS context. It is perhaps stupid but I don't think it should be considered improper. The fairness philosophy is a good one, not because others do it at 300 nanoseconds, but because it is a good thing.

At one level the 350 microsecond delay, thanks to the magic shoe box, is the equivalent of making the IEX system a co-location facility the size of NJ. That is, just a slow exchange for all. However the differential use of faster data within IEX, such as with DPEG and their routing, makes it quite a bit different in some regards.

Where I think there should be traction in the voice of disapproval is that IEX is essentially a slow exchange. IEX provided numbers in their application and you can go and look at the cute graphs. I don't like the idea that slow exchanges should be approved and gum up the system but there is no regulation against it. Reg NMS is certainly not really designed for protected quotes coming from Chicago nor the West Coast, let along for slow exchanges near NJ/NY like IEX. In Reg NMS there is verbage of the form that as long as something happens in the last second things are OK. A whole second. How quaint. There are 1,000,000,000,000 picoseonds in a second. Reg NMS needs some revision.

There has been much talk around having a minimal latency threshold, such that platforms shouldn't be allowed that slow the NMS down, but that regulation doesn't exist. I personally think something harsher should exist so that if you are below the current median performance of all exchanges you shouldn't get a guernsey on the basis that you are not improving the system, you're making it worse. That is not a thing. I also think there are too many exchanges and they should be more rigorously controlled but that is also not a thing. These are problems with IEX's application but they are not IEX problems. They are problems with Reg NMS and it would not be fair to exclude IEX on this basis.

Reg NMS has resulted in a pretty good system. As good as the NMS is, it has issues that could be improved upon. The bottom line is that there are worse existing problems to be found than those in IEX's application.


In summary, IEX should be allowed to be an exchange subject to the B/D conditions and without being allowed protected quote status. However I don't think this is possible within the current regulatory framework. Such an approval is likely to be unlawful. I guess we'll have to wait and see. The people at IEX are good people and deserve the status they are seeking so let's hope they can make some headway in reforming the system eventually.

After all, it is wrong to deny people the good grace of capitalistic failure.

Happy trading,