Of course, this resonates with what the great historian Arnold Toynbee said about why some civilizations thrive and others fail. After a lifetime studying societies spanning 6000 years and five continents, Toynbee wrote that the one common thread appeared to be whether both leaders and the people chose stodgy obstinacy or agile flexibility, whenever challenges loomed. And especially whether they gave support, invested resources, and enthusiastically backed-up their creative minorities.
And hence, this time we'll peruse a potpourri of science marvels showing that agility and scientific creativity have not become endangered species -- despite the efforts of some at both political extremes. Indeed, we're still displaying an eagerness for pragmatic problem-solving may yet help us to thrive.
Let's start with this interesting news. GM has demonstrated an energy storage system built from five used Chevy Volt batteries, which would be capable of providing two hours of backup power for three to five average homes. As the companies note, while they're no longer suitable for use in an electric vehicle, the average end-of-life battery still retains about 30 percent of its charge, which can go a long way in other applications (especially when a few of them are linked together). Of course, this is all still just at the demonstration stage, but I am already interested! I'd love to have a cheap version to charge with a used solar panel... just enough to keep my fridge running for a few days of blackout. There's a real commercial potential there. Hey GM, need a celebrity spokesman?
==Geosciences and the Earth==
First some very mixed good news. The boom in availability of natural gas in the U.S. from shale formations is not without (fracking) controversy. But it means the North American price of methane is less than half of what it is in Europe. In a boost to the U.S. economy, manufacturers have plans to invest as much as $80 billion in U.S. chemical, fertilizer, steel, aluminum, tire and plastics plants, according to Dow Chemical. And the main reason, said George J. Biltz, Dow Chemical’s vice president for energy and climate change, “comes back to the massive competitive advantage the United States has with natural gas today. One can hope that economic recovery will then allow calm people to start picking more carefully which areas to subject to these new processes and carefully supervise the professionalism of the frackers... and choose to protect sensitive realms that they must avoid.
These changes will also be geo-political, as U.S. imports of oil from the Middle-East have actually started to decline, reducing American dependence and...perhaps shifting our security focus. This in turn may affect political balances... and it will undermine the grip that coal has on the current economy. Since methane procuses half as much atmospheric carbon per unit of energy as coal, and much less of the ancillary poisons, this is guardedly good news, providing we not let this slow down our drive to research even better methods. Speaking of which...
Next year, when the California Ivanpah desert solar plant flips the on switch, it will nearly double the amount of solar thermal energy produced in the United States. According to Dr. Steven Koonin (my old Caltech classmate and recently under secretary of the U.S. Energy Department) solar thermal is the most under-rated sustainable energy system around, with great near-term potential for profitability, even despite cost pressures from the plummeting cost of natural gas.
The World's largest offshore wind farm is coming online. The U.S. could have been the world leader by now, if the first decade of the 21st had not been wasted. But yay for this. And it's not too late.
Now let's swing toward the energetic... but weird! It's possible that, at the microbial level, the deep seafloor is humming with current. Danish researchers have found bacteria that conduct electricity along microfilaments from the sea bottom's surface to many centimeters down beneath. With so much electricity being transferred, are other organisms tapping the lines? Might the Desulfobulbaceae be a power source for entire as-yet-unappreciated deep-sea microbial ecologies, which in turn shape some of the planet's fundamental biogeochemical processes? Hm... did I hint at this in EARTH?
== Ocean Fertilization: Right idea... wrong guy ==
In July, a rogue entrepreneur named Russ George dumped 100 tons of powdered iron into currents off British Columbia. The intent: to trigger plankton growth and aid in the recovery of salmon fishing, while also removing CO2 from the atmosphere. Marine scientists have termed his action “unscientific, irresponsible and probably in violation of international agreements.” A foolish stunt, indeed!
Yet, some of the outrage went too far. In Testing the Waters, Naomi Klein rails against any form of geo-engineering experiments, even those that mimic totally natural phenomena, the same phenomena that create the world's great fisheries that feed a third of the planet. (70% of the oceans are mostly-dead deserts. But fertilizing updrafts off the Grand Banks, Chile, Antarctica etc create fecund, oceanic oases.) As an equal opportunity contrarian, I call on the left to back off a bit. Most of what Ms. Klein says is true... yet I find her polemical reflex is unhelpful and possibly toxic to our future.
That kneejerk reflex is to assume that technology-based solutions are automatically suspect, possibly evil, and that any palliation of the thing they are complaining about will reduce the need or desire or imperative to eliminate the problem at its source. That is illogical, self-righteous and lazy thinking, in the extreme. We need to be examining and dispassionately studying palliative measures both because they may be our last resort... and because they may help us transition, even if we apply our main efforts to doing the wise thing and stop befouling our planet.
In other words... I support limited, small scale ocean fertilization experiments that mimic natural phenomena by expanding the realm of life. They are, in essence, no different than irrigation that we do on land. Rife with potential problems, but a winning scenario, if done carefully.
Having said that, let me add that in the specific case in question, I think it was a doltish, oafish stunt, in the wrong place and the wrong time. And illegal to boot. But you can expect more such experiments in the future, under the protection and auspices of countries like Nauru that are threatened by rising oceans. And, if I lived in such a place, I would be investing in better versions of the idea -- like wave-powered bottom stirrers to bring up natural sediments, more closely imitating the natural updrafts off Chile and the Grand Banks. (I depicted the method in my 1989 novel EARTH.) And I would tell Ms. Klein to go turn her ire on Fox and the Kochs, but practice a little humility and patience toward those who agree with her that the world needs to be saved! They just want a backup option. A Plan B.
==Biosciences & Medical Advances==
In 1773, when Benjamin Franklin's work had moved from printing to science and politics, he corresponded with a French scientist, Jacques Barbeu-Dubourg, on the subject of preserving the dead for later revival by more advanced scientific methods, writing:
I should prefer to an ordinary death, being immersed with a few friends in a cask of Madeira, until that time, then to be recalled to life by the solar warmth of my dear country! But in all probability, we live in a century too little advanced, and too near the infancy of science, to see such an art brought in our time to its perfection. (Extended excerpt also online. )
Wow... what a guy. Preserved in Madeira? Okay. Unlikely, but a pretty great image.
== The future is fun! The future is Fair! ==
Did anyone notice the "More Science" part of this posting's title? I recently had the good fortune to meet Phil Proctor of the Firesign Theater. Brought back memories of Bozos and Nick Danger and... more science!
Which somehow segues into this: neuroscientists identify how zebrafish regenerate brains and other organs after trauma. (Yeah, back in the 1960s I often needed brain regeneration!)
And more! Yay science! German brain researchers have successfully induced Tourette's syndrome symptoms in healthy people for the first time, using powerful magnetic pulses. Oh yeah? Well f#@k ain't that f#@king great?
== And it's Science Miscellany time! ==
A prototype ultra-sensitive sensor would enable doctors to detect the early stages of diseases and viruses with the naked eye.
Tiny 3-D printed bio-bots will crawl through your body, targeting toxins.
The protein folding problem has been a Grail of biology some time. Now a team claims they can predict how one will loop and fold, in advance. A big deal.
And some health advice: Cool your palms and build muscles and lose weight? Heads up to keep an eye on this. Exercise does more for you if you cool your palms and the soles of your feet? Huh. Some of you write in and tell your results.
Oh and I hear that healthy young adults ages 18–25 can improve their working memory by increasing their Omega-3 fatty acid intake.
What we die of: A graphical look at the primary causes of death in 2011.
Germany is set to advance a bill Wednesday imposing a spate of new rules on high-frequency trading, escalating Europe’s sweeping response to concerns that speedy traders have brought instability to the markets. As I have said, this may be more important than anyone as yet knows.
This is kinda neat. The BioLite stove burns regular wood or twigs etc to cook with... but also generates electricity for a charging unit. Volunteers took several into areas blacked out by Hurricane Sandy and made so many friends their sales are booked into next year. Great for the next disaster…or that unforeseen Zombie Apocalypse.
Rollable-foldable electronic devices! As predicted in EARTH (1989) and in EXISTENCE!
Tentacled robot mimics the movement and capabilities of a soft-armed octopus.
A small but growing cadre of savvy technologists argue that, at least in measured doses, encounters with imaginary worlds and futuristic devices could have a decisive influence on innovation. David Brian Johnson, Intel’s staff futurist, even insists in a recent book, Science Fiction Prototyping, that by writing stories about future products, engineers can do a better job of actually making them.
It will likely take a decade, but improvements to lithium-ion batteries could lead to much cheaper electric vehicles.
A new approach to create panoramas from live camera feed on mobile phones.
A whopping 100,000 entangled photons have been detected for the first time, beating the previous record of just 12. The technique for spotting this delicate quantum link among so many photons could prove useful for safely sharing keys used in encrypted communications.
The Counter-electronics High-powered Microwave Advanced Missile Project (CHAMP) renders electronic targets useless, a “non-kinetic alternative to traditional explosive weapons that use the energy of motion to defeat a target,” CHAMP emits bursts of high-powered energy, effectively knocking out the target’s data and electronic subsystems. Most press reports have incorrectly described this as an electromagnetic pulse (EMP) weapon. High power microwave (HPM) is a different technology that uses a microwave beam that can be focused tightly to hit designated short-range targets.
How cool is this? Reminds me of George Gamow's Mr. Tompkins in Wonderland. MIT video game lets you play with relativity, changing speed of light.
==Science and Society==
The U.S. used to by far have the highest ratio of college grads, but that is hard to maintain while absorbing half of the world's immigrants. Funny thing though, the country that is now number one, with 51% of adults having a degree is the other great land of immigration, Canada. The US is still a very respectable 4th place. Actually pretty amazing, all considered.
Especially after hearing that science literacy is actually improving and apparently because of those sappy "breadth requirement" science survey courses that non-science majors are required to take in U.S. universities (but not, apparently, in most European or Asian colleges.). In fact, because of those few college breadth requirements, the US scored first in adult science literacy! Of course, one could argue whether this applies to all the different Americas, red or blue or...
Still. Take that you cynics. As for the rest of you, keep plugging for more science!
While I was still with my last employer (CAISO) they were listening to some of the stakeholders in the electricity market and they were talking about some very interesting possibilities for those used batteries. If the vehicle fleet gets large enough there are two things that become possible that don’t work well today.
Vehicles can be charged overnight increasing the base load during normally low demand periods. If that stored energy is then placed back on the grid during peak hours, it can change the reliability strategies used by utilities. Energy suppliers currently overbuild in order to provide for peak power demands. That extra production capability increases their operations costs and that gets passed to retail customers. Some of the peak producing units also use higher cost fuels (especially true in areas not blessed with abundant hydro power), so our retail rates get hit twice. Anything we do that increases base load demand that prevents peak load demand is generally a good thing as a result.
Splitting production and consumption requires energy storage and that has been the holy grail of the industry. One of the US East Coast ISO/RTO groups changed their market rules to enable this stored energy to be sold back through aggregation services and might still be doing it. If you live in such a region, not only do you benefit from an averaging effect on your electricity demand, you might be able to profit directly from the price differential. Any market trader can easily imagine what that could do to the price differential. Even second hand batteries could make a dent in this decades-old opportunity.
Don’t expect auto batteries to be worthless when a car is ready to be junked. The utilities should be interested in buying them up. In fact, the utilities might wind up being the major buyer of old vehicles or their end-of-life battery packs. We might get a step closer to cradle-to-cradle manufacturing where our industrial processes chain sinks and sources together. Doing that for energy will be a game changer.
For natural gas, I suspect the price differential will finally force the US to figure out how to export the stuff. Our ports are going to change.
I love the Franklin quote. If the connectomics people are right, he could have had a sporting chance -- had he only gone through with it! Luckily there are a few others scattered throughout history whose brains were chemically preserved, such as Einstein and Babbage.
Franklin has another interesting tie to the cryonics movement as well, as a pioneer in the area of long-term "Methuselah" trusts. (Not every attempt was so successful -- reference the cautionary tale of Jonathan Holdeen)
Currently doing an online Stanford course on solar power, batteries and fuel cells (as well as, harking back to last post, settling a newly acquired puppy)
It has been shown that, in principle, a distributed mix of solar thermal and wind farms could provide all of Australia's stationary energy needs.
The feed-in DC-AC inverter being a substantial energy sink, and since the first thing a good many electronics components do is switch AC back to DC, I wonder why you can't have panels providing a direct DC outlet for powering appliances? (Old Chevy batteries included)
More Science High! Shoes for Industry. Shoes for the Dead...But who really runs Communist Martyrs High School?
Can anyone tell me what happened to the typePad Architechs wiki page? Used to be at:
It'd be nice to have a regular Wikipedia page about the show, which was way cool and should not be forgotten
adiffer... interesting. Already there's a huge push to get bus and truck fleets to go natural gas, those on reliable routes.
The future is fun
The future is fair
You may already have won
You be already BE there!
Hope you use the Paglia on Star Wars response in your blog. I do have a question about the use of iron to boost plankton drawdown of CO2. Won't this just increase the acidification of the oceans, which in turn could destroy the shells of phytoplankton and crash the food web? Is there any geoengineering process to reduce ocean acidification?
So, I betook me to the Hashfire Inn for a secret caucus of the hotheads.
Rebellious libertines all, lusting for life and liberty.
George Washington brought the hemp, and I the evening papers.
We quickly proceeded to get Sam Adams and young Tom Jefferson goodly stretched by the hemp, which smoked us all like Boston scrod.
What a fetid fervor of freedom!
“I say, let’s have a revolution!” said fiery Sam the tax collector. To which Big Ben replied,
“Fine Sam, then we can invite over a bunch of immigrants and make cars."
But even before Firesign was Beyond the Fringe with this...
Followed the link to your piece about high frequency trading, David, and your proposal for a transaction tax to effectively end the practice. By way of full disclosure, I am an algorithmic trader myself, but I think you presented a pretty strawman-ish version of the case for HFT, and you seem ignorant of one particular improvement over the old status quo that it offers which I think ought to be very appealing to you.
- The main argument in favor of HFT is not some ideologue's incantation about "efficiency," but rather the fact that it has forced market makers into quoting narrower bid/ask spreads, which *reduces* transaction costs and the level of "parasitic gain" being extracted by middlemen in the market. Crossing a 25 cent spread to buy a stock represents an immediate and significant up-front cost to enter a trade. If the spread is only 1 cent wide, the investor pockets that difference. And it is *only* via the fast reaction times made possible by HFT that market-makers are able to quote such narrow spreads. This is why a seemingly unnoticable 0.1% transaction tax could actually impose far higher costs on investors.
- The part about HFT which you might find particularly appealing is that it has opened up the field of market-making and made it much more democratic. Becoming a specialist on the floor of the NYSE or other major equity exchanges used to be a matter of knowing the right people and being born into the right New York families. Now, all it takes is a colocated server, an account with adequate funding, and some programming and trading savvy. Granted, this is still a significant level of resources beyond most people's reach, but the field is now far, *far* more open to greater numbers of players than it used to be.
- Positing a causal link from the correlation between the market crash of 2008 and the rise of HFT is quite spurious. The crash was caused by a runup in bad debts from mortgage securities and exotic derivatives, the vast majority of which were not even traded on listed exchanges. HFT firms did not generate systemic risk and at no point during the crisis did an HFT outfit require a bailout (unless it happened to be owned by one of the mega-banks). Trading volumes have been declining in recent years, but that's an outcome consistent with a burst asset bubble and a 12-plus year sideways bear market and moribund economy. Anyone who thinks HFT is to blame for this should go back and take a look at the collapse in stock market volumes in the 1930s, before anyone on Wall Street had even heard of a computer!
I do think there are some potential problems with the practice, some of which you pointed out. However, they can all be addressed with pragmatic and targeted solutions, rather than a transaction tax, which amounts to the proverbial sledgehammer swung at a mosquito:
- Yes, HFT algos can sometimes sniff out large orders and attempt to trade ahead of them. (This is, by the way, a market making practice as old as the markets themselves, certainly much older than the advent of computerized trading.) The solution for this is, first and foremost, development of better algorithms by the institutions placing large orders, to more effectively conceal their intentions (also a very old practice). This process is already well under way.
- A much bigger problem, as I see it, is the "arms race" aspect of HFT, where huge resources are devoted to IT infrastructure meant to wring the last few nanoseconds of latency out of a trading system. This does result in a kind of rent-seeking for whomever can create the fastest system, and isn't necessarily very socially useful. (Although I would argue that the vast sums spent by the financial sector on IT resources throughout the process of computerizing finance has at least given a respectable boost to technology R&D in the U.S.) Chris Stucchio, who runs an excellent blog discussing, discusses some possible solutions to the arms race problem here: http://www.chrisstucchio.com/blog/2012/hft_whats_broken.html
Oh, and the possibility of an HFT firm creating an AI system which runs amok seems *extremely* remote to me. Generally speaking, there's too much of a short-term focus at trading firms for them to undertake the kind of long-dated, significant investments that would be required to build genuine artificial intelligence. The task that they're ultimately trying to accomplish (buy low & sell high) also doesn't really lend itself to spontaneous emergence of AI, in my opinion.
Things go wrong, and trading systems can and do spin out of control sometimes. But (as the recent Knight Capital debacle demonstrated), there's a limit to how much damage they can really cause to the market when they do. Rapid loss-making algorithms are limited to blowing up the balance sheet of the firm that runs them (and they're a boon to pretty much everyone else in the market). A theoretical very fast gain-making predatory algorithm (no such thing has yet been invented) would be limited to taking money as fast as other market participants would be willing to lose it. In fact, I think this latter limit effectively makes a highly predatory AI trading system impossible to build.
In the last thread, Ian Gould mentioned
"a novel set in a galaxy divided between alpha Humans and Beta Humans. The Alphas are our descendants - the Betas are the descendants of humans taken as servants by aliens centuries ago."
I don't know if it's the one he was thinking of, but James Hogan's Giants trilogy deals with conflict between Earth humans and those who live off-planet with the "Giants" of the title.
Andrew Bissel wrote:
"The main argument in favor of HFT is not some ideologue's incantation about "efficiency," but rather the fact that it has forced market makers into quoting narrower bid/ask spreads, which *reduces* transaction costs and the level of "parasitic gain" being extracted by middlemen in the market. Crossing a 25 cent spread to buy a stock represents an immediate and significant up-front cost to enter a trade. If the spread is only 1 cent wide, the investor pockets that difference. And it is *only* via the fast reaction times made possible by HFT that market-makers are able to quote such narrow spreads. This is why a seemingly unnoticable 0.1% transaction tax could actually impose far higher costs on investors"
This strikes me as misleading. Yes, it is true that HFT allows for "narrower bid/ask spreads", but it is not at all obvious that this provides any real benefit for investors. For an investor (that is, someone who buys stocks and holds them for some period of time in the hope that s/he will earn money over the longer term) as opposed to a trader (someone who is profiting from arbitrage or some other aspect of the trading itself) the difference between a one-cent or a two-cent spread is almost certainly irrelevant, and quite possibly lost in the noise of overall transaction costs.
The actual benefit of HST seems to accrue to traders, who can profit from very high-volume trades via arbitrage or in the space of bid/ask spreads.
Indeed, HST may well increase the level of "parasitic gain" (in total, even though reducing it in percent per trade) by enabling far more high-volume, high-speed trading in the spread space. This can result in vastly more trading, but the volumes would seem to benefit only those doing the trading, with little or no benefit to any normal "investors".
An EV charging system might be configured as an energy cache, conditioning power and leveling demand. businesses might even wish to offer charging at cost, for the benefit of a parking lot full of potential back-up power. Not the first time GM powered a house, in the 60s they provided all the electricity, heat and air conditioning for a large house with an Olds V-8, claimed it cost less to run than the normal utilities. These days it could be done with vastly less engine, on natural gas. And if the late Thomas Gold was correct, fracking should work almost "Fracking" anywhere. And I want to see more solar, OTEC, and geothermal to reduce the needed holes in the ground.
We're talking about a reduction of imports from not only the Middle East, but from Alberta, Canada as well. This may end up defusing the fears of eco-crises and political unity fractures in Confederation over pipelines such as the proposed Keystone XL and Pacific Gateway projects...with perhaps less relief from the unity fears over the latter, as it's aimed at sending product across the Pacific to clients on the Austral-Asian side of that ocean.
The possible loss of American clientele for Albertan Tar Sands oil does seem to have kick-started some interest in refitting, repurposing and expanding existing infrastructure for sending the oil eastwards across Canada for domestic sales and consumption, however. Whether or not that's a good thing over any timeframe you care to discuss...?
As for Naomi Klein, while she's not any direct acquaintance of mine...
I've no doubt that she's already been working for some time on the problems posed by the actions and attitudes of the Koch brothers and Fox News. She might have been shocked - with some cause - to see experiments such as this one you've reminded me of taking place much closer to her old homestand here in Canada, though.
I await the results with as much interest as either of you, with reasons of my own for it.
Andrew Bissell, thank you for your response from the perspective of a person intimately involved in the world of High Frequency Trading. I appreciate the fact that you come to us both intelligent and sincere. You are certainly welcome here and this brilliant blogmunity hopes you'll hang around.
Nevertheless... having said that... (and I'll break up this reply into pieces)....
You also assert and style yourself to be scientific and offer up polysyllabic incantations in hope of convincing us that is so.
It is not so. Sorry. It is diametrically opposite to so. I have seldom read such unscientific... stuff... as your missive contains.
First, you demonstrate that you either did not carefully read my article or simply did not understand it. Especially regarding the theromodynamics of life. You illustrate this by saying: "has forced market makers into quoting narrower bid/ask spreads, which *reduces* transaction costs and the level of "parasitic gain" being extracted by middlemen in the market."
This is towering nonsense. The very worst aspect of HFT is that it eliminates nearly all of the free energy/enthalpy gradient and profitable difference available to living market participants.
Sighing, I will try yet again to illuminate the bloody obvious.
In nature, the bio-energetic range from plankton to Killer whale... or from grass to lion... is NOT composed of a murky, flat slurry of ten million tiny gradations of predation.
From sunlight to grass to lion there are three, maybe four large steps, because only such large steps make the staircase usable. Such large increments make the hunt worthwhile. They are essential to maintaining large, complex metazoan organisms. Tae staircase from high enthalpy sunlight to re-adiated infra-red waste heat must be taken in large chunks.
When the grass and the antelope and the Lion are suffering from parasites, THAT is the equivalent of HFT, flattening the steps and robbing the lion of his profit.
You can wave all the mumbo jumbo incantations that you wish, but nothing takes away from that parallel in nature. Or in engineering where it is the accessible DIFFERENCE in temperature between two heat pools that allows a generator or motor to operate.
You manage to hypnotize yourself into believing it a GOOD thing that you eliminate the gradient and macroscopic difference between the value of a security perceived by the buyer and the seller. But in so doing, you deprive the winner of that bet from gaining anything worth his time and energy spent in the "hunt."
You are absolutely and 100% wrong. Diametrically and spectacularly wrong... and illustrating the power of self-hypnosis.
Andrew Bissell: next item:
Oh but then you go on to: "The part about HFT which you might find particularly appealing is that it has opened up the field of market-making and made it much more democratic."
I stared at this remark... then stared at it some more... and found it impossible to believe you would actually say it. That anyone would try to assert (forgive me) such an astounding sophistry. Oh, I understood the incantation. I simply could not grasp how you can say such things with a straight face.
Dig this, please. HFT works ONLY for those players who have the staggering advantage of being seated exchange members. If they had to pay even a tiny commission, they would have to stop. This monstrous system serves the insider only.
I did not blame HFT for the 2008 collapse. That was primarily due to OTHER behaviors of your microcephalic employers, who all style their modes of parasitism as "genius." Oligarchies have always done that. The low IQ Kaiser and Czar Nicholas each flattered themselves as geniuses, just before setting the world afire and destroying the great hopeful world of 1910 and then torching their own dynasties. Thus, too, the seated members think that they control this tiger.
Nevertheless, HFT has been involved in some pretty awful mid-level calamities that showed the danger of relying on automatic systems that are developed in utter secrecy, not subject to the normal reciprocal criticism of scientific civilization. Before proclaiming "no worries" show me when that process has ever, ever ever worked well or had a happy ending. One example will do for now. Just one.
They will get worse.
Andrew Bissell: final response (and please forgive the rapid, first draft quality; I have very little time):
Your blithe assertion that systems controlled by the seated members of the exchange can somehow be kept blind to the intentions of outside traders ... is delightfully charming!
Outside investors who are compelled by archaic "seating" rules to USE the seated members as middle men... these investors' orders must cross into the castle of seated members... and you believe these comings and goings can be kept secret and perfectly confidential, when there is zero synnergy to ensure it at all?
Speaking to you as an expert in such matters (The Transparent Society) let me ask you once again: can you cite for me an example of such assurance that has ever proved valid across decades? EVER? This naive fantasy should qualify you for membership in the science fiction writers of America.
As for the TERMINATOR scenario, while colorful, it is also highly plausible. The fact remains that HFT is the branch of AI research that currently gets the most funding and is undertaken in the greatest secrecy, under motivation rules that are utterly predatory and parasitic in goal and action.
You shrug these facts aside and claim they cannot lead to malignant AI because HFT is highly specialized.
Again, you exercise tunnel vision. OTHER areas of AI research are dealing with the problems that HFT is too specialized to cover. Vision systems, language, and so on, these are proceeding apace elsewhere, developing modules that are not secret and hence are available to be plucked up whenever a proto-AI chooses..
In Intelligence theory, one key hypothesis is that AI will emerge when some kernel, somewhere, achieves the speed and voracity to latch onto the diverse elements that are now dispersed -- from google's self-driving car to IBM's Watson. That Kernel need not be intelligent in its own right. What it needs is lavish resources and fantastic speed and lightning access to the world. And zero public accountability.
All of these are traits being poured into HFT by men who are blithely confident in their own sagacity and who deny any need to subject their self-serving assumptions to critical review.
I have to say, that I am disappointed. I had hoped that you would offer us something, anything, to justify this horrific trend that only benefits a seated oligarchy, enticing them to abuse an archaic trading system that should be open, without need of entitled club members. Members who can do a billion trades without fee, while charging investors who want to do just ten.
That you would defend such a blatant interference in trade shows you are no child of Adam Smith.
But do come again! ;-)
Hostess Executive Bonuses: Twinkie-Maker To Seek Approval For $1.8 Million In Bonuses During Liquidation
if'n y'all don't want a class war, you should stop poking it with a stick.
Andrew Bissell, I had thought that one of the recent market bobbles (several hundred drop in the Dow) was due to an out of control trading algorithm:
These algorithms are meant to take advantage of microsecond fluctuations in market share values, that are unavailable to us peons who buy and hold, even if only for a few weeks. That's why some big trading houses are moving as close as possible to the actual floor of the Dow, just because shaving off the milliseconds it takes in electronic transit to place an order to buy or sell can result in hundreds of thousands in profit.
704 Begnize: Home of the International League of Red-Headed Gentlemen
danger of relying on automatic systems that are developed in utter secrecy, not subject to the normal reciprocal criticism of scientific civilization. Before proclaiming "no worries" show me when that process has ever, ever ever worked well or had a happy ending. One example will do for now. Just one.
Alright. I'm not sure what the threshold for "automatic" is. How about our nuclear program. It was developed in Secret, to the degree that neither the public in our nation, nor (IIRC) the leaders of our allied countries knew we were working on the Atom bomb. The program didn't have any 'automatic' features at first, but it does today, and those features were developed in secret. It seems to have turned out well so far, with only a couple nukes ever used in war.
Sociotard: Yeah. That Hostess-related item smells not unlike trying to bait the bear into a rampage. They may not believe they're doing any such thing, but "reasonable observer" standards might suggest otherwise.
sociotard, anyone who thinks that even a secret military programs is "not subject to the normal reciprocal criticism of scientific civilization" does not understand the military system. In the US that culture is obsessed with red team challenges. Moreover, they were frightened by simulations like FAILSAFE, DR STRANGELOVE, ON THE BEACH and so on and policies were changed accordingly..
I'll be getting back to Ocean Fertilization and other science-related topics in a tiny bit. In the meantime, I thought this article concerning the surprise that Romney lost might be of interest to certain parties... especially given a certain host's odd belief in conspiracy theories about vote switching systems. Of course, if we learn in the next year or two that the FBI or NSA has done an investigation into Diebold and found irregularities... and seized control of the systems to prevent their being switched, I'll print out my words and eat them. Without condiments.
Still, it is a weird amount of blindness here. Why did BUSINESSES blindly believe Obama would lose? They're risk-adverse by nature. Wouldn't you expect them to hedge their bets? Though I suppose... some did. Those who stuck with the Democrats at least. ;)
I have to admit that I don't mind importing stuff from Alberta. I think we can afford to watch patiently while you work out the political issues. Heh. Maybe the Dakotas will try to sucede and join you all instead. 8)
Tony: It's all about voltage drop. If you have the PV's on your rooftop, the devices closest to them see the highest voltage. We do the conversion thing as a technique for reducing long range losses, but if you run your own array you can produce something similar by running lines at a higher DC voltage.
Look at the voltage on your old land line phone (if you still have one) and you'll see it is pretty high. That's how they ensure it will work well enough through the entire house.
So you're saying the military engages in sufficient internal criticizm of its secret projects, but the business world doesn't?
Being someone other than Dr. Brin:
Considering the reaction to Obama Volume II, I have to believe that certain factions of the business world are not yet willing to resume self-critique.
Sociotard, you're talking about private and public sector differences. The private sector tries to maximize profit by minimizing effort. The public sector doesn't care and instead spends the extra money to try and make it work right. That, and the military was probably worried if something DID happen, it would blow up in their faces. The financial sector is of the "what, me worry?" mindset and would gladly see the world burn... after they made their profit and retired.
The unaffordability of college in the US is going to start dragging that science literacy rate back down.
I will say that the survey courses I took in each of the major sciences when I went to college provided me with a *fabulous* grounding in science. Everyone should do it if they get the chance without going into debt slavery.
To put some concrete numbers on the cost of the bid/ask spread, back in the late 1990s before decimalization and computerized market-making took off, it was not uncommon for stocks on the NASDAQ to have 25-cent wide spreads. If you assume that the true value of the stock was right in the middle of the spread, then an investor who traded "at the market" (lifting the offer to buy the stock, and hitting a bid to sell it) overpaid by 12.5 cents each time he entered and exited a position. So, say that you correctly guessed that eToys would rise from a value of $50 to $60. Your actual profit from the trade would be 59.875 - 50.125 = $9.75, meaning 2.5% of your gain was lost to paying the spread. Now, not all spreads were that wide, but this shows you how the spread can be relevant even to long-term investors. And if that doesn't seem like all that big a bite out of the profit to you, consider that those costs were being paid to the MMs by everyone trading in the market, regardless of whether they lost or made money.
A transaction tax or a speed limit like the Germans are implementing will not necessarily force spreads back to that kind of width, but it will make them incrementally wider, and the costs from that will be shared between regular investors, and the market-makers (who'll take a hit from the reduced trading volumes it's likely to produce).
It's true that an incremental improvement in the spread of a few fractions of a penny probably isn't realistically relevant to long-term investors. But doesn't that also indicate that HFT (at least the vanilla market-making variety) isn't likely to harm those investors either? I guess what I'm asking is, if a tightening of a few cents in the spread isn't a point in HFT's favor, why is scalping a few cents (the usual accusation against HFT) considered so harmful?
"In the US that culture is obsessed with red team challenges. "
Like Millenium Challenge 2002?
Top level military culture in the US is stultified, rotten, and doctrinaire. Read _The Generals_ for how it *used* to be.
I think your biological analogy has a flaw. Maybe you could point out your thinking on this. Yes… I’ve ready your ‘trading monster’ post. I think you are missing an important point in it regarding anti-fragility and in that you are unwittingly advocating for a limited set of parasites who feed upon us.
I was taught to respect the role that parasites play in the wild. A decent measure of the health of an organism is its freedom from parasites, but when measuring the health of the environment in which you find that organism it is best to measure the diversity of the parasites that would attack. The more diverse the parasite population, the more likely the organism they feed upon is to be healthy. No single organism is ever truly free of parasites, but the successful ones manage to adapt toward symbiosis with some of their would-be killers which is enough to stress the parasites that don’t adapt as well. When the parasites are diverse, many more options are available to an organism and solutions become a little easier to find through trial and error.
I view some of the HFT's as parasites on the parasites and some as parasites on those of us on the outside. Sometimes they are both, but I'm not that upset by their presence because they come in various shapes, sizes, and levels of lethality. If they are all using the same algorithms, colluding to limit the competing parasites, or hiding where we can’t respond to them it is obvious we have a problem. Rather than eliminating them, though, I'd rather there were more. I would rather take the biological analogy a step further and increase their diversity. I recognize that most of the cells in my body don’t have my genetic code in their nuclei, yet I am a descendent of a long line of people who have thrived. They had to deal with parasites of many types… and did.
I’m not making a market efficiency argument. I’m pointing out the advantage that Taleb describes as anti-fragility. A parasite-free organism cannot adapt to stress with as many options as one that is a collection of organisms. We are anti-fragile in the sense that we can get stronger with each stress because we adapt within each of our components. Our economy is such a collection. Obviously it would be more pleasant if we faced less stress, but I think it is pretty obvious that stress motivates adaptation.
As long as we keep pushing for transparency, I’m optimistic we can beat the $^%!!# by adapting to our real environment. We are members of the greatest civilization ever, after all. 8)
Andrew: you "spread" argument for HFT is pure, unadulterated BS.
First of all, spreads are *still* 10-15 cents, routinely. HFT is not eliminating large spreads!
The change in spread is actually due to *decimalization*. Previously, with eighths the minimum possible spread was 12.5 cents, so a 25 cent spread was only two ticks. (Decimalization was an excellent idea.)
Second, you don't seem to know what spreads actually *are*. They are *NOT* fees paid to market makers; that's just bullshit.
If I make a limit order, the limit order sets a bid or ask price, and simply sits until the bid and ask cross each other and other order settles.
The market maker gets paid nothing, unless he decides to go ahead and take my order.
And thanks to HFT, it is *completely unsafe* to place market orders; it is now a truism in the investing world that only suckers use market orders, because if you use a market order, the HFT scammers will front-run your order and give you a worse price than the real market.
This is the key point: market orders are no longer used by any serious investors, thanks to HFT. HFT has destroyed the market order.
You clearly don't do actual investing, Mr. Bissell. I do. HFT is a front-running scam which has destroyed the usability of market orders. It should be stopped.
The transaction tax would be an actual cost to investors, but it would be a cost paid very little by long-term investors, more by day traders, more by speculators, and mostly by the HFT scammers. In short, it is a Pigovian tax which taxes the most socially destructive behavior. It's a good idea.
Ah, I see I didn't read your full disclosure, Mr. Bissell: you aren't an investor. You're an algorithmic trader -- in other words, you're talking your book.
HFTs caused the flash crash. Seriously, isn't that enough to ban them? I don't particularly care that you're going to lose your shirt. *Nothing* you are doing is socially useful.
The social purpose of investing is driven *entirely* by long-term investors, and we long-term investors don't need ANY of the rest of you in order to fulfill the social function of investing (allocation of capital to useful projects). Of course the presence of "traders" is literally as old as stock markets, and probably unavoidable, but it doesn't mean we should be concerned about their welfare.
Why should we care about the demise of the 'market order'?
Lots of great dialogue here today.
I too dislike iron in the ocean for exactly the acidification issues. As a non-expert I nevertheless suspect that a source of CaOH used in conjunction with iron might make a beneficial net contribution, but we don't have pure CaOH available without CO2 generation... so it remains a somewhat pointless hypothesis. A calcium meteor might do sudden magic... a game changer.
I would ask anonymous posters sign some name at the bottom of their posts because we have too many anonymous posters.
(Hmm, I once knew a guy who just signed "sock puppet" but even that helped keep a conversation going.)
Madiera probably would not have enough alcohol to keep Ben viable until some latter day Victor Frankenstein came along. Although as a prelude to an enjoyable afterlife it has some merit.
Rum would be a better choice and actually was used for such purposes, usually with a death at sea.
Admiral Nelson was supposedly thus preserved after he fell at Trafalgar. The phrase "tapping the Admiral" for "let's have a drink" is a later invention, spawned by ghoulish but unlikely tales of thirsty tars helping themselves to a few sips of Horatio's preservative on their way home.......
Perhaps a tot of that would give you the strength of eleven men.
Your replies were actually much lengthier than I was expecting! Arcana of trading isn't necessarily the most interesting subject to most people, so I appreciate your having taken some of your limited time to respond. I'll break up my replies as well.
To put the self-apologia first, you're incorrect in your assumption that I work for a big bank or other organization that was engaged in the kinds of dealings that led them to demand bailouts to save them from their bad bets. FWIW, in 2008 I was of the opinion that they should have been allowed to fail. I'm independent now, but when I started in the industry I only worked for small proprietary trading firms, which bear their own losses and don't pay out bonuses when the group loses money. To an extent I understand why people sometimes jump to lump all of finance under the category "Wall Street" and condemn it in those terms, but it still makes me a bit prickly since I earn my keep without the benefit of taxpayer backstops, even while competing those who do get those subsidies.
Okay, on to the meat of your responses:
W.r.t the parasite metaphor, I guess I'm a little confused as to exactly which activities in the market you consider parasitic, and therefore exactly how the market lines up with the biological systems you described in your other post. However, if you'll permit some gross overgeneralizations, market participants can be broken into two broad classes: market-makers (MMs) and directional traders (DTs). DTs include your classic long-term investor, who thinks a stock's undervalued and wants to buy it one day and sell it for more a few months or a year later. MMs are very different. Within their relevant timeframe, market-makers are indifferent to whether they buy or sell a given security, as long as they can, on average, buy it for less and sell it for more than it's worth, and in order to compete with other MMs they at least occasionally must be willing to pay a bit more or sell for a bit less than everyone else in the market. The money they earn from trading that spread must be sufficient to compensate them both for time & effort and for the various risks they incur (first and foremost, that a position will move against them while they hold it in inventory).
Market making is not unique to HFT. DTs have always paid MMs in order to enter and exit their trades. The main difference is that a computer-based MM is able to quote a tighter spread because it can react more quickly to changes in the market, and thereby mitigate its inventory-based risks in ways that human traders can't. So the key change that HFT has wrought has been to force most of these human MMs out of the market, replacing them with a larger number of algorithms competing for smaller amounts of profit per trade.
On parasites, cont'd:
This all takes place in a realm of competition largely irrelevant to the longer-term DTs. (By way of example, here's Jim Chanos of Kynikos, a long-term investor if ever there was one, pooh-poohing the idea that HFT has much impact on his business: http://video.cnbc.com/gallery/?video=3000117254)
So, are the old human MMs who've been displaced the "lions," and HFT is reducing the profits that can be made from trading the spread to an unhealthy level? This was the idea I had in my mind, and since I see quoting unnecessarily wide spreads as a form of rent-seeking, that was why I said that I believe HFT is actually reducing a form of parasitism in the markets.
Or do you mean that HFT threatens to take too many bites out of the DTs' profits, to break them into too many finely-grained transactions? I agree that would be a problem, but I don't think it emanates from the MM side of HFT. It's true that not all algo trading programs are MMs. Some seek to exploit statisical relationships or trends in the market over long periods of time. But the longer the timeframe, the more indifferent they are to latency, so I don't see how they're relevant to a discussion of transaction taxes and other measures to curb high-speed trading.
adiffer you make a cogent parallel... but only so far. We do not have an ecosystem, in which the roles are well tested and defined. We have one filled with untested invaders. HFT is not like the common cold, constantly testing us and making our immune systems stronger. It is a parasite that is insatiable, that kills hosts even in violation of its own self- interest.
Tacitus, I'd want Ben F's brain no matter how dissolved. ANY added data would help us to model him.
Andrew Bissell your reply to Greg clearly demonstrated that you are buying into a voodoo incantation. That the purpose of the market is to efficiently zero in on the mystical "right price" for a security. And that narrow spreads are a good thing.
Let me be plain. You have absolutely no basis for that religious catechism. It is unsupported by thermodynamics or engineering or by any kind of science at all. Certainly not by economics!
The crucial thing at play in securities markets is the DIFFERENCE in expectations between the buyer and the seller, This is what maps onto the the thermodynamic steps I described earlier. Of course, the buyer and seller each think they are right and one of them will turn out to be the winner-predator. But dig this. If micro transactions evaporate this difference, there will be no incentive for the buyer to buy or the seller to sell. Except for those for whom this is a game, the MM folks...
And the MM folks are not what equities markets are supposed to be about, which is the allocation of capital in the real world.
By flattening the gradients so that ONLY expert manipulators can make a living by eking slivers off the edges of tiny-myriad trades, you have not created free enthalpy. You have created entropy and death.
Thanks for your distinction between DTs and MMs. Alas though, those distinctions are chimerical compared to the real distinction that I referred to again and again, without you apparently bothering to read what I wrote. (So why should I bother?)
That is the distinction between seated members of the exchange and everybody else. Seated members may trade without penalty (commission) meaning they can engage in HFT. No one else can do so. This is an inherent monopolistic-cartel practice, even were HFT to have no other bad attributes. Moreover, your failure to address this proves that you have not read the things you are commenting on.
THAT is what I hold against you. You've been courteous and engaging. But the fact that you do not read what you comment on, that is unacceptable and a sign of a character flaw. The flaw of incuriosity.
If you want to compare HFT to a theoretical market structure that has never existed, where any hobbyist has the exact same level of access to order flow and the same power to trade as exchange members, then I'd agree that today's status quo doesn't seem very democratic at all. If, on the other hand, you compare HFT to the old specialist system that preceded it, and what the markets would likely return to if HFT were completely eliminated, I think it is undeniable that the newer way of doing things is far more democratic. There are more market makers competing on a greater number trading of venues. By being colocated they have a fraction of a second's edge in being able to react sooner to publicly disseminated data feeds, but that is miniscule compared to the advantage that the human NYSE specialists used to enjoy in being able to observe and react to all the order flow as it made its way to the floor.
None of which is to say that HFT and electronic trading hasn't actually made the field more open to hobbyists. I know a guy who a few years ago wrote to the Interactive Brokers API and put together profitable algo trades that he ran in an account with $25,000. He had to compete over longer timeframes to overcome his disadvantage in not being colocated, but he did make money. That has become harder to do as more entrants come into the market to exploit those opportunities, but it is still possible.
At the same time, the playing field has also become more open for those wishing to become seated exchange members. This varies by market and by exchange, but in many cases the expense of the technology required to be competitive now dwarfs the cost of exchange membership dues and colocation fees.
It's true that the market-making activity in some of the most liquid, highly competitive stocks is now dependent on liquidity rebates and the absence of commissions, but this is not at all true of all HFT market making. I am not a member of the derivatives exchange where I trade; I pay fees both to the exchange and my broker for every trade and receive no rebates, but I'm still able to turn a profit using HFT methods.
Just speaking personally, I know that I would likely never have become a trader if not for the existence of HFT. Over the decade that it came into being it changed the profile for a typical exchange market maker from "member of the Yale club, college football player, aggressive bully" to "skilled, nerdy programmer and market geek." Scoff if you like, but I call that pretty damn close to "democratization."
You're not the only busy man on this thread, David. I'm composing my replies as quickly as I can while remaining thoughtful and extensive and attending to other pressing matters in my life.
Andrew Bissell your reply to Greg clearly demonstrated that you are buying into a voodoo incantation. That the purpose of the market is to efficiently zero in on the mystical "right price" for a security. And that narrow spreads are a good thing.
Your accusation that I haven't read your posts or Greg's is false. Just because I'm not reaching the same conclusions about the market structure as you are does not make me "incurious" and that frankly strikes me as a bullying tactic on your part. Though not an effective one, because here I am, staying in the fray!
The problem you're describing -- where in the near term relevant to market makers all trading happens at "dead value" (as we in the business call it) -- isn't at all new to me. I've seen precisely that thing happen in a market where algorithmic trading became more common and more competitive. But as far as I can tell, the only people it has harmed have been the old, hands-on human traders that used to be the ones scalping the spread, and who couldn't compete with new systems. A long-term investor, wishing to express a view in that market over several weeks, months, or years, can still do so, at lower cost of entry & exit, and in most cases with greater amounts of liquidity available for him/her to take.
Or is it your contention that HFT operating in milliseconds is somehow dampening volatility in the market even over the very long term, making trading in those timeframes not worth pursuing? If so, I don't really see how that's the case. AAPL's about as algo'd-up a stock as you can find, but if I had sold it back at $700 and bought it today I would have made plenty of money from the trade.
I'm trying to understand your point about HFT moving the markets toward an entropic elimination of the exploitable differences in price that make trading a market worthwhile to buyers and sellers. Over the extremely short timeframes relevant to HFT I agree that this can happen. In fact, I'd say it's probably the proximate cause of the steep drop in profitable that HFT firms are witnessing right now. There's been a classic overinvestment in capacity, and it's putting the squeeze on everybody in the business. But that process is self-limiting; it can only proceed so far before the marginal competitors will get squeezed out of the market, restoring enough of the enthalpy needed to support the remaining players.
Will try and address some of the other matters ("calamities," AI risk, etc.) as soon as I can.
You'll be hard pressed to find the day that Knight Capital blew itself up on a chart of the stock market with labels removed from the date axis. The Dow did not drop several hundred points. The impact was mostly limited to less liquid names where Knight was the primary market maker, where -- sure enough! -- spreads widened out until some relative normalcy returned to the market.
That one of the top 3 equity market makers could blow itself up and have to withdraw from trading, with relatively muted impact on the general market, is actually a pretty good demonstration of just how resilient the market is now to those kinds of shocks.
Some further (and much better discussion) here: http://www.chrisstucchio.com/blog/2012/flash_crash_flash_in_the_pan.html
Andrew, you claim to comprehend the topic you are discussing here, yet you never address any of the fundamental points.
Proclaiming that HFT is more democratic, when only seated cartel-members can engage in it, is almost orwellian in its staggering degree of reversal of truth.
You say that seated memberships are cheap to acquire. Really? Aren't they inherently limited in number and supply? Indeed, jealously guarded? I ask this not out of hostility but genuine curiosity? If the supply is unlimited and low in entry price, then in theory the cartel nature of the seating discrepancy might go away. Is that the trend? Toward a day when every e-trade person can have a seat on the NYSE?
If not... and let's be clear... there is not a scintilla of justification for calling HFT anything but an oligarchic obscenity.
Even were the inherent cartel barriers not there, you said it yourself. HFT is inherently favorable to the big boys by reason of access to super-expensive tools.
As for the distinction between DTs and MMs, you actually seem to believe the nostrum that the MM dance of narrow trade margins helps the world of commerce and products and services. I have studied the arcane incantations that you fellows use to make this claim and it is absurd. There is one trade that helps a company to capitalize... the issuance of new stock.
"it changed the profile for a typical exchange market maker from "member of the Yale club, college football player, aggressive bully" to "skilled, nerdy programmer and market geek." Scoff if you like, but I call that pretty damn close to "democratization."
Oh I will agree with you there! Quants and geeks are now able to stand up face to face with Harvard frat boys. Is THAT what you meant by democratizing? I'd be impressed if the Quants were managing the economy better, but they aren't. In any event, though I come from Caltech and know and identify with geeks. I have NO sympathy for your assertion. Trading one set of secret masters for another is unappealing.
You appear to be trying. But I honestly think you keep appealing to what most of us can clearly see to be quasi religious catechisms.
One other quick point, while I have a little time:
... the distinction between seated members of the exchange and everybody else. Seated members may trade without penalty (commission) meaning they can engage in HFT.
Sadly this distinction, and the insider advantages enjoyed by exchange members, will not at all be eliminated by elbowing out HFT through a transaction tax or other measures. In fact, they will probably be amplified.
The only way to eliminate this distinction would be to start an exchange where there was no membership and by definition no insiders. But I don't think it would attract any market makers, and without the liquidity they provide, trading on that exchange probably wouldn't attract other investors & traders either. I'd love to be proven wrong though!
No ecosystem has roles that are defined and tested. By definition, they are vulnerable to disruption by an innovator. The order that emerges (if it does) is of the spontaneous variety that can be incredibly fluid and complex like fractals or trivially simple. Only the designed orders are testable as they tend to have limited complexity that can be enumerated and evaluated. I don't want to live in a designed economic order.
Ponder the possibility that this is a bit like your transparency argument. Getting rid of the parasites and even limiting them is like getting rid of the cameras and microphones. It won't happen. We are better served by having more of them pointing every which way. I'll accept some reasonable limits, but I think we are better served when the population of parasites is diverse and able to feed upon each other to some extent because we can compromise with the less lethal ones.
I understand that the parasites kill. Even the regulars like influenza can be quite deadly. We can adapt by understanding what they do and adjusting what we do using that data. A transaction tax on the HFT folks might limit them, but we are still exposed to others. How confident are you that your idea would produce a better situation in this ecosystem after the other parasites adapt to our action? My confidence in our ability to foresee these things is near zero.
An ecosystem is a poor analogy, the stock market is a contraption a tool, it is intended to funnel capital to new enterprises.
That is it's function - and its benefit to society
Adding features that do not help it's function - or worse make it function worse is a bad idea - like carrying an anchor in a car - or like one of those idiots who try to fit a wind turbine to a car
Given that people will try and vampire from the stock exchange a transaction tax to discourage the more blatant forms of abuse is a good idea
A transaction tax could have another advantage,
To me the stock market looks like a machine running with no damping - a bump causes excessive reaction, this sometimes kills productive companies
A transaction tax would take money out of the system proportional to the rate of transactions - just like a viscous damper in a car suspension system
This should reduce the massive reaction to small inputs
Andrew. You sadden me. I had thought that you were saying that the seated member cartel was self-dissipating. News that I had not heard. Alas, you reconfirmed the rigidity of the cartel... then made the next fantastically opposite-to-true statement:
"Sadly this distinction, and the insider advantages enjoyed by exchange members, will not at all be eliminated by elbowing out HFT through a transaction tax or other measures. In fact, they will probably be amplified."
Sorry, it is nonsense. The seated members can engage in free infinite trades and can do HFT. Non-members of the cartel cannot. Period. Absolute and totally absolute fact. Every non-cartel member experiences a huge transaction tax called "commission" plus the delays of using middlemen who are less than trustworthy.
Seated members have no such transaction tax. You are defending the utterly indefensible.
adiffer I agree that an "ecological" accomodation might be found with parasites as happens in nature. But that is during equilibrium. We are NOT in equilibrium in the economy or equity markets.
New methods that are monopolized by a cartel are inherently predatory and parasitic and cannot be commensal or symbiotic. The core of transparency is reciprocality and an open, fair playing field.
Andrew, as Nathaniel points out, your "spread" argument is based as much or more on the former part of the "before decimalization and computerized market-making" than the latter.
But I think you give your game away when you say "an investor who traded "at the market" [...] overpaid by 12.5 cents each time he entered and exited a position." After all, an investor will only infrequently enter or exit a position; someone who is constantly entering and exiting positions is trader (whether an individual or a large firm).
I don't have the cite, but I recently read that, for 2011, some 70% of all trades were HFTs. I submit that the "harm" comes from skimming off the wealth into the hands of the high-frequency traders, whose activities provide no actual benefit, even to investors.
Your blithe assertion that systems controlled by the seated members of the exchange can somehow be kept blind to the intentions of outside traders ... is delightfully charming!
That's not what I said. I said the *primary* measure to mitigate the risk of algo trades front-running large flow will be making it harder for them to detect it. A big part of that involves the institutional investors developing their own algorithms. But that's not all that needs to be done. I'm also open to reasonable regulatory suggestions to deal with this as well. The Themis Trading guys (big critics of HFT) have pointed out that counterparty trade tagging allows an algo to determine where a large player wishes to buy or sell a big chunk, and it can then place orders to trade just ahead of that limit. Maybe that should be banned or restricted in some way. Note that this is one potential unfair use of HFT that might not be effectively remedied through a transaction tax or speed curbs. Ultimately there will be some boundary to how effectively we can limit the ability of market participants to suss out large interest and trade ahead of it (big trades into relatively less liquid markets leave their own footprints), but I'm sympathetic to the argument that it has become too easy.
As far as the potential for a rogue AI to escape from a trading company's systems and run rampant goes, I won't delve too much into that topic because my knowledge of AI is probably around 1/10000th yours. My doubt that it would arise from HFT is just a matter of doubting the ability of my fellow program traders to come up with code that sophisticated. Simply put, where it comes to programming, I think we're good, but we're not that good. The only times I've heard AI mentioned in my industry have been as a buzzword in sales pitches for trading strategies that, when you look under the hood, are always quite straightforwardly algorithmic, with very little in the way of machine learning incorporated. But if an HFT firm actually has advanced the state of the art in AI in any significant way, I'd be very interested to read about that.
You may argue that we haven't heard about it because HFT firms are so secretive, but not everything they do is held so close to the vest. See, for example, the now open source LMAX Disruptor architecture. Also, my admittedly cagey sources at the big firms like Getco and Citadel haven't ever mentioned significant use of AI. In fact, as near as I can gather the trend at many of these firms is in the direction of less complicated algorithms, for instance, gaining a latency edge in parsing market data and generating orders by loading all trading logic onto FPGAs. Can we safely aver that a computer architecture with no concept of memory isn't likely to become self-aware?
But I think you give your game away when you say "an investor who traded "at the market" [...] overpaid by 12.5 cents each time he entered and exited a position."
Don't get hung up on "each time." The 2 trades made to enter and exit a position are sufficient to establish the case. (Although I'm tempted to prod you to try and draw the line between "investor" and "trader." Obviously "constantly" entering and exiting positions is impossible. But what about the guy who turns over his portfolio, say, 3 times a year? 5 times? See where I'm going with this.)
Decimalization only creates the potential to quote at 0.01 spreads, but no credible observer believes it would be possible to make markets that tight while limited to human reaction times. Reducing the ability to quickly cancel a resting order, by definition, increases the risk associated with placing that order in the market, which has to be compensated by wider spreads. One can credibly argue that that the cost of those wider spreads is somehow offset by improvements it makes to other aspects of the market ecosystem. Even improving investors' perceptions might be one such mechanism for it to do that. What one cannot credibly argue is either a.) that speed curbs will not result in wider markets, or b.) that tighter spreads, ceteris paribus, are not a boon to longer-term investors.
By way of an olive branch, here are a few issues (in addition to the colo cable length I mentioned above) with HFT that I think regulators *should* address:
- It has become a common practice for many HFT firms to get their brokerages to waive the normal pre-submission order risk checks. Those checks take time and add latency, so as soon as one shop convinced their broker to suspend the check, all their competitors requested the same thing just to keep the playing field level. I think some kind of standardized risk checking interval, required before submission of any order from any trader or dealer, would do a lot to mitigate the risks of large errors and blowups like we saw with Knight, while keeping the playing field level for all entrants.
- There's very little outside supervision of the leverage being extended by brokers to HF firms. I suspect most are pretty well-capitalized (risk departments at most brokers are usually looking for any reason *not* to extend special trading privileges), but the devil is in the details, and it would make sense to check and be sure that they're not granting 10x leverage to firms carrying open position risk for extended periods of time.
- It's possible that the steep losses typically suffered when a firm unleashes an errant algo are not enough of a deterrent. So it might make sense to levy fines against firms whose error trades have a noticeable impact on market price. (I'm less sanguine on this one ... I'm so loss averse that the penalty meted out by the market seems more than enough to me. But it's possible there's some kind of risk calculus going on at the larger firms where they've determined that letting an algo occasionally spin out of control is less costly than developing the controls to prevent it.)
- There's currently no cost to jamming a lot of quotes into an exchange with the sole objective of slowing down the market data feed. I'm not sure of the best way to address this. The CME uses a message/volume ratio. I think they set theirs way too low, but something like that might be the best solution.
Anyway, that's probably the last I'll jump in here, but thanks to everyone for the interesting & enlightening dialogue, and particularly to David for playing the occasionally-sharp-elbowed-but-gracious host. :)
To return to the original topic.
1. I doubt that the mpact of a few million tonnes of iron on the oceans' acidity would even be measurable compared with the impact of billions of tonnes of carbon dioxide. (Plus, of ourse, the addition of the iron is suppsoed ot promote the removal of dissolved Carbon dioxide (AKA carbonic acid. So you'd expect a net reduction in acidification.
2. As the news gets ever direr, it's plain we need to throw everything at the problem: emission reductions, adaptation measures; seqestration at source and from the ambient atmosphere and geo-engneering.
I don't think there's a geo-engineering fix.
I do think, geo-engineering my keep the msot disastrous effects at bay for an additiona; decade or two to allow us to develop other solutions.
3. I'm surprised that adding aerosols or particulates to the upper atmosphere - the geo-engineerign technique the National Academy of Science found most promising - doesn't get mentioned more.
4. Another idea i'll toss out: has anyone considered large-scale cloud seeding over te arctic regions during winter to increase snow and ice accumulation?
On the ocean fertilization topic, I don't believe it was illegal. He apparently did it in Canadian waters and the Canadian government knew about it.
It turns out Canada is not a signatory on the geoengineering treaty. So he did not break any laws.
Andrew Bissell, I found one thing rather disconcerting: your view of long-term investors. Long-term investors are NOT people who buy with the intent to sell. Long-term investors are people who buy and then use the dividends of the stock as a source of income. These are people who buy into a company and remain a stable core. Sometimes it fails them - there are lots of stockholders who got burned at Hostess (along with stakeholders such as my friend who lost his job and his pension while executives are getting millions in bonuses). But there are others who do quite well by keeping money in a few stocks for years and years without the intent to sell but instead of utilizing dividends.
What you describe are short- to medium-term investors. These people buy stocks with the intent of selling them for a profit. These are people who do not intend on keeping the stock but instead want to make a specific profit.
So. When we say a transaction tax will not harm long-term investors... it won't. Anyone who already bought a stock won't be touched. Anyone buying the stock after it's implemented will be hit once or twice. But people who buy with the intent of selling and then rebuying and reselling... these will be hit by the tax the most... and they're the ones using the most infrastructure and causing the most instability in the market.
Plus, maybe they'd get real jobs and increase the GNP, and provide the rest with actual useful stuff.
Andrew Bissell I have explained the thermodynamic and other reasons why "efficiently finding the correct price" -- the dogma-catechism justifying MM style trading -- is utter hogwash. That, plus the brutally blatant unfairness and market distortion of the seated-member cartel, is more than enough reason to diagnose that Wall Street is desperately sick.
HFT is an amplifier of all sins. It is fuel poured on each and every fire.
But all of that is said and I'll not belabor the point and I don't want to drive you away. You have potential for being a valued new member of this blogmunity. So let me switch to a question:
You said there were two types of equity trading activity.
MM which is (I interpret) rapid trading with an aim of nibbling at small valuation differences that may bear no relation to the products, services or management or prospects of the company that long ago issued the equities in question
DT which (I believe) means investors who are making wagers on the rise and fall of the extended value of that company based upon.
But you left out a third category and I found that omission quite telling. Because it is the diametric opposite of MM. It is what the equity markets should be FOR, the infusion of fresh capital into productive enterprises. It is the Issuance of New Stock.
That third category is the only one that actually serves productive enterprise. It dilutes the shares of current owners, but supposedly with the intent of infusing capital to do more business and increase the company's overall value. And stockholders are welcome to buy some of those shares, often at an owner's discount.
Sorry Andrew, but your emphasis on MM is diametrically opposite to the function of markets that actually helps productive enterprise. So opposite that you never even MENTIONED the third option.
In fact, all cap gains from trades that do not infuse capital TO companies should be taxed as regular (gambling) income, that benefits nobody by gamblers like you...
... but newly issued stock should get a 10% rate for 5 years. THAT would bring new capital to companies and increase the value of ALL stock as innovation and profits rise.
I will try to come back here once or twice, but we have a new posting.
In fact, all cap gains from trades that do not infuse capital TO companies should be taxed as regular (gambling) income, that benefits nobody by gamblers like you...
... but newly issued stock should get a 10% rate for 5 years. THAT would bring new capital to companies and increase the value of ALL stock as innovation and profits rise.
I'm actually in favor of taxing capital gains at a level equal with income (would prefer that it be accomplished by an offsetting reduction in income/payroll taxes so as to be revenue neutral, though, but that's my "government takes enough in taxes already libertarian" side showing). I'm engaged in providing a service and I should be taxed accordingly. The tax rate I pay actually is already pretty close to the standard income tax, a function of the markets I trade and the fact that all my trading is short-term and not subject to some of the carried interest deductions that are often used.
To your other point, about the "third category" of investors buying into IPOs or secondary offerings, I guess I see this as being subsumed under the category of the directional traders/investors. Presumably, someone who buys stock when it is first issued hopes to gain from that investment through a combination of price appreciation and dividends, which will happen over the longer term (setting aside, for the moment, the IPO flippers and the MMs). The dividend vs. price appreciation thing is kind of a distinction without a difference; they are both methods by which a publicly traded company can deliver value to shareholders.
It seems like everything that happens in the stock market aside from an IPO or secondary issuance is under assault here as "speculation" and "gambling." So let me point out: the existence of an actively traded, liquid secondary market for their shares allows companies to raise far more capital in exchange for a given share of their company, because IPO investors will pay more if they can be assured of being able to realize gains on their investment when they choose, with minimal losses due to illiquidity. This is what makes issuing stock attractive in the first place -- otherwise, companies would just avoid all the regulatory burdens and reporting requirements that come with it. Those reporting requirements hint at another major social good contributed by the stock market: it widely disseminates information about the financial health and operations of public companies. (This was true even before the SEC standardized the public reporting requirements.) The books aren't always trustworthy, but for the most part, the balance sheet and income statements of a publicly-traded company are far more transparent and easily acquired than those of private, closely-held firms.
I had put up a post earlier addressing your questions about the barriers to entry in HFT but it seems to have been lost in the ether. I'll try to recompose and post it within the next couple of days.
Thanks again for the discussion.
"What one cannot credibly argue is either a.) that speed curbs will not result in wider markets, or b.) that tighter spreads, ceteris paribus, are not a boon to longer-term investors."
You continue to conflate 'speed' and 'frequency', as you do 'computer trading' and HFT. These two pairs are not twins. One can easily engage in fast computer trading without doing any high-frequency trading, for example by enabling one to more quickly respond to a bid or an ask.
Note further that a minimal financial transactions tax would have minimal impact on such fast trading. It would fall heavily only on those who are engaging in actual HFT: making vast numbers of trades back and forth in the attempt to squeeze a tiny amount out of each, but leading to a large gain overall.
So far as I am aware, no one it suggesting "speed curbs"; what some are suggesting instead is a transactions tax, which would likely function as a frequency, but have no effect on 'speed' itself. No one would pay a higher tax because they could execute a transaction more quickly.
Further, even if we were to grant that "tighter spreads, ceteris paribus," are "a boon to longer-term investors" (something that I submit is not at all well-established, at least at the level here under discussion), it is plainly the case that things are very much not "otherwise equal" in the current discussion.
 That is: some level of very large spread might begin to affect normal investors in some way, but even the "large" spreads mentioned in the discussion here are, for an ordinary investor, sufficiently small to be "lost in the noise" of commissions, transaction costs, etc.
Andrew I have little time and can't come back to old postings for long. You can always follow me to the latest comments section at http://davidbrin.blogspot.com/ where our lively community resides.
I do not mind your proposal that most of tax reform be done on a relatively neutral basis. That is... AFTER the top 2% are asked to pay more, period, for reasons of oligarchy-prevention. (And you really, really need to read Adam Smith.)
But to show you how I generally agree with you on principle, see http://www.theamericanconservative.com/articles/revenge-of-the-reality-based-community/
As for your refusal to recognize the existence of a third equity trading regime, I am afraid that I'm simply boggled by the obduracy of perspective that your community has created in your mind. You can honestly tell us that you see no difference between DT, buy-and-hold investment based on value growth... versus the issuance of new stock? Really?
Might I ask you to step back for a moment and... follow the cash? Follow the ACTUAL MONEY! DT is honest wagering while MM is parasitic nibbling. But both are gambling casinos that have almost no bearing upon the health of the companies involved. Their only "benefit" to say IBM or Apple is to adjust the price/value... so that IBM or Apple might thereupon gain fresh capital by issuing more stock! (Or by using stock as collateral for bonds.)
Issued new stock... the money goes to the company! It can be spent on equipment and research! None of the gambling profits in MM or even DT go to such things. Not one bit. Thus it is NEw Stock that should get favorable tax treatment, helping persuade owners to allow a little dilution in exchange for increased company value.
It is definitely a "third type" of stock trading and the only type that merits any sort of favorable treatment. DT is neutral, owners exercising their rights. MM is flagrantly and insanely harmful, justified by a voodoo that claims narrowing the enthalpy gap is GOOD for the parasitism victim. It is not.
But I will back off from everything, if you will help us pass a law forcing all seated member stock chairs to be split into a hundred seats and ninety-nine of them sold off. The seated cartel-oligarchs would get cash for those 99 spun off seats! And we would all see the end of an anti-competitive, conspiratorial, dreadfully dangerous cartel.
My wife Ramona and myself have been continuously off of the electric grid for 222 consecutive days today. I designed the combination solar and wind system, which also relies on 33 KW of deep cycle battery backup, to meet ALL of our electrical power needs, which it has. This is quite achievable for homeowners. I anticipate reaching 365.25 days off grid in due course.
Firesign Theater?!! Nick Danger third eye etc!? My Gawd reminds me of my misspent high school years. Hint, while these albums were funny no matter what, after smoking a bit of MJ...death from laughter induced asphixiation was a very real possibility...
Post a Comment