http://www.wired.com/business/2012/0...treet_trading/
One article of many. What're your thoughts? Does this practice rely on inducing extra volatility? Is that a good idea, considering the potential for error?
Printable View
http://www.wired.com/business/2012/0...treet_trading/
One article of many. What're your thoughts? Does this practice rely on inducing extra volatility? Is that a good idea, considering the potential for error?
It does increase day-to-day and minute-to-minute volatility tremendously, even during "normal" operations -- not the buggy kind.
It's a good idea for companies like Knight Capital.
I'm not particularly worried about volatility or the occasional large losses in a small firm. I do wonder, though, what the actual utility of spending an absurd amount of money on a few extra ms is. Obviously it's an arms race now, so it's a perfectly rational decision. But in the broader point of providing liquidity for the market, I find it likely that we're approaching a point of diminishing returns.
The reason these companies make money is because they're exploiting inefficiencies in the market - often they have information and can act on it faster than the rest of the market, so they have a definite advantage. The only way to take away this advantage is for exchanges to intentionally introduce friction into trades, a minimum 'floor' for trade execution time. Then the arms race would shift back to the best algorithms rather than the fastest links, and probably save a lot of money.
That being said, I'm very leery of this kind of solution. People might have felt the same when traders started shifting to computer-driven trading rather than voice-driven trading, but obviously the ancillary benefits are quite significant beyond merely exploiting some timing inefficiencies. Who knows what clever advantages/products/markets might come out of the current wave of innovation?
http://www.latimes.com/business/mone...,2381214.storyQuote:
Knight Capital Group received the emergency funding the major Wall Street brokerage needed to stay alive, according to financial media reports.
CNBC reported two private equity firms, General Atlantic and Blackstone, lined up with brokeragesTD Ameritrade and Stifel Nicolaus to invest $400 million in the beleaguered Jersey City, N.J., firm.
Ever since Knight disclosed Thursday that it had lost $440 million because of a software glitch that sent a slew of errant trades into the stock market, the firm has struggled to stay alive. Knight reportedly needed and got a lifeline Friday to keep it operating through the end of the week.
A Knight spokeswoman declined repeated requests to comment Sunday.
If the deal has indeed been struck, it would avert potential failure of a Wall Street powerhouse. Knight, founded in 1995, processes some 10% of U.S. stock trades.
Algorithms vs speed won't matter much unless there's better programming and quality control. IMO
Apparently Knight approached the Securities and Exchange Commission (SEC) for a quasi bailout by canceling all the trades. The SEC turned them down. Good for them. :up:
Software screwup. It happens, and the consequences are real. The solution isn't banning software.
Dread, who said anything about banning software?
Quote:
... quants—the physicists, engineers, and mathematicians-turned-financiers who generate as much as 55 percent of all US stock trading.
Quote:
No sane human trader would spend their time haggling over a ten-thousandth of a cent, but computers don’t get bored.
The quotes in Yahoo’s order book probably came from an algo, and you almost certainly can’t trade at that price. Even if you had access to the exchange, which of course you do not, they would likely be gone long before you could jump in the market—either already executed or, much more likely, withdrawn before any shares changed hands. And that’s where the really dangerous part begins to set in. It’s not just that humans are less and less involved in trading; it’s that they can’t be involved. “By the time the ordinary investor sees a quote, it’s like looking at a star that burned out 50,000 years ago,” says Sal Arnuk, a partner in Themis Trading and coauthor of a book critical of high-frequency trading titled Broken Markets. By some estimates, 90 percent of quotes on the major exchanges are canceled before execution. Many of them were never meant to be executed; they are there to test the market, to confuse or subvert competing algorithms, or to slow trading in a stock by clogging the system—a practice known as quote stuffing. It may even be a different stock, but one whose trades are handled on the same server. On the Internet, this is called a denial-of-service attack, and it’s a crime. Among quants, it’s considered at most bad manners.
Quote:
High-frequency trading raises an existential question for capitalism, one that most traders try to avoid confronting: Why do we have stock markets? To promote business investment, is the textbook answer, by assuring investors that they can always sell their shares at a published price—the guarantee of liquidity. From 1792 until 2006, the New York Stock Exchange was a nonprofit quasi utility owned by its members, the brokers who traded there. Today it is an arm of NYSE Euronext, whose own profits and stock price depend on getting high-frequency traders in the door. Trading increasingly is an end in itself, operating at a remove from the goods-and-services-producing part of the economy and taking a growing share of GDP—twice what it did a century ago, when Wall Street was financing the enormous industrial expansion of the economy. “This is counterintuitive, to say the least,” wrote New York University economist Thomas Philippon in an article for the Russell Sage Foundation. “How is it possible for today’s finance industry not to be significantly more efficient than the finance industry of John Pierpont Morgan?”
I have high regard for the quants and math geniuses, but this kind of HFT is perverting the original intent of stocks and exchanges -- as a way to lend businesses capital for expansion, and earn a little profit for having invested in their growth. This is math and technology making money off math and technology, using speed and velocity and volume that's got nothing to do with the underlying real assets.
Some proposals are mentioned here: http://www.bloomberg.com/news/2012-0...ace-rules.html along with comment suggestions to Tax each transaction. I still like the idea of somehow separating exchanges: one for high-risk market makers, non-participating speculators, and HFT math quants, and another for "retail investors" where fundamentals of buying/holding/selling can still work.
Either that, or find a way to attract these quants into fields like....making personal jet packs to replace cars, or killing cancer cells at the speed of light, or.....
Uh, your idea about having separate exchanges is terrible.
Why terrible? It's not much different than buying mutual insurance instead of a shareholder for-profit insurance, or using a credit union instead of a commercial bank. I also like the idea of a structural "firewall" between retail and investment banks, and between insurance companies and investment firms *cough AIG*
edit
wiggin, there are already plans underway for local/regional exchanges. They didn't seem to think it was such a terrible idea....
http://online.wsj.com/article/SB1000...647926882.html
No it's not. It has very little effect on that "original" sort of trading. It can be largely ignored by anyone in the market who isn't trying to make their money by constant trading, HFT has next to no effect on long-term performance, your "fundamentals of buying/holding/selling".Quote:
Originally Posted by GGT
Wow, you just mixed so many things up in one post it's hard to know where to begin. Local exchanges are to trade stocks of local businesses and raise small amounts of capital outside of the big exchanges, where they'd never get listed and have lots of issues. They are NOT to locally trade stocks of large companies, which would just be illiquid and never get used because of bid-ask spreads, trading costs, etc.Quote:
Originally Posted by GGT
Reimposing Glass-Steagall provisions and similar things may not be a bad idea, but that's completely unrelated to having a separate exchange for 'retail investors'... and I don't even get where you think credit unions and mutual insurance come into this.
The reasons your idea is terrible is because the end product for said retail investors would almost certainly be worse than the general market, and it would make price discovery much more challenging.
You may not like the way I described "different exchanges", or maybe the comparisons to banks/insurance wasn't slick enough, but the gist was reducing risk/volatility by removing certain actors that cause added risk/volatility, and bubbles. It's (sort of) similar to separating house-flipping speculators from conventional buyer-owners, by tweaking tax codes and length of ownership, putting 'investor-buyers' in different tax brackets. Mutual insurance companies do business for their policy owners, not shareholders.
HFT has made price discovery harder, at least for the average retail investor. Even when economies are tanking and employment is a problem, stocks brokers are doing fairly well, seemingly disconnected from the real economy and traditional investing. minx's article confirms some of that. none of this shit happens in a vacuum
GGT, the underlying prices between two exchanges can't get too far out of synch or you get arbitrage. You'll have the same risk of long term bubbles in either exchange, and short term volatility is not a big issue for the buy-and-hold investor; liquidity, cheap trading, and small bid-ask spreads are far more important.
Your comparison to taxes related to housing and mutual insurance is yet another non sequitur.
As for price discovery, I don't think it means what you think it means.
...she said before posting an anti-software rant and suggesting that software trades should be segregated. :downcast:Quote:
Originally Posted by GGT
Then I'll try to make my points clearer. The article wasn't just talking about Knight Capital, but more broadly about electronic trading. And not just the digital part that enhances 'market efficiency', but how math quants and trading on speed alone can impact the financial parts...in negative ways. Those physicists, mathematicians, and computer geeks (who add 55% to volume of trading) likely don't know much about the underlying assets themselves (like a financial advisor or economist would) but are working in the ether of math and speed (like a NASA engineer).Quote:
Originally Posted by wiggin
Nano-trading to exploit a process, time and technology, in order to make money --- that isn't really investing. In essence, we've got a financial/economic system (in the big board exchanges) that's being used as a technology system, with regulators and oversight always two steps behind. When investment firms like JP Morgan hired 'the first quants' to create synthetic derivatives -- CDOs and MBSs -- they were so complicated hardly anyone understood what they were, or how they would change 'the markets'. Not even the CEOs or fund managers, let alone the SEC. That's why they were called Weapons of Mass Destruction, and a handful of folks warned against them, and at the pitfalls of ending Glass-Steagal.
Now that we've seen the fall-out from those derivative WMD, today's warnings come from HFT. The flash crash a while ago wasn't just a fat-finger one-off....e-glitches are more frequent, and involve multiple millions of dollars. Facebook's IPO and Knight Capital are just two examples.
Anyway, IMO this is one reason behind consumer confidence flagging. It explains why the stock market has a low 15% "approval rate", with banks around 20%, and feeds the narrative that the system is rigged, corrupted, broken, more like gambling instead of investing, etc. The little guy, retiree, retail investor is thrown into the swamp with big fish, HFT, day traders, speculators and hedge funds. They can't compete or keep up, but their money is still swimming in the swamp. Billions get taken out and put into Treasurys or bonds, as flight to safety, turns into earning losses (can't even outrun inflation). Even parking cash in a bank doesn't yield enough interest to keep up. Everyone knows it's a crappy way to save or invest, but there aren't many alternatives. Especially for older geezers like me, who don't have a 40-50 long term time frame like you.
I think millions of people like alternatives and choices. Real choices, not just default positions. Congress and regulators naturally move slower than innovation; therefore, I'd like to see more newly created options. Local or regional exchanges, a different way to invest in municipalities (maybe Social Impact Bonds ? ), maybe a National Bank or an Infrastructure Bank. That's where I was going with my reply (and mentioned credit unions vs commercial banks, and mutual insurance vs publicly traded insurance). It's feasible those things will happen before credibility, confidence and trust are restored to big banks or big boards.
You mean you would trust someone with science and math on their side about how an equity's price is likely to change over a financial advisor with a hunch? Seems counterintuitive to me.Quote:
Originally Posted by GGT
No one ever said it was investing, we all agree it's trading. You're straying from the question of separate exchanges into a broader complaint about regulation. There may be some worthwhile arguments in there, but they're irrelevant to your suggestion about making two separate exchanges.Quote:
Nano-trading to exploit a process, time and technology, in order to make money --- that isn't really investing. In essence, we've got a financial/economic system (in the big board exchanges) that's being used as a technology system, with regulators and oversight always two steps behind. When investment firms like JP Morgan hired 'the first quants' to create synthetic derivatives -- CDOs and MBSs -- they were so complicated hardly anyone understood what they were, or how they would change 'the markets'. Not even the CEOs or fund managers, let alone the SEC. That's why they were called Weapons of Mass Destruction, and a handful of folks warned against them, and at the pitfalls of ending Glass-Steagal.
You have yet to prove that high frequency trading or algorithmic trading significant changes the long term price of an equity, and that's all that matters to the retail investor. This isn't a question of 'competing' - the losers to HFTs are day traders, not the long term investor. The long term investor gets significant benefits and likely fairly little in the way of costs. You have yet to address the very big problems I raised with having separate exchanges and how it could possibly benefit a retail investor.Quote:
Anyway, IMO this is one reason behind consumer confidence flagging. It explains why the stock market has a low 15% "approval rate", with banks around 20%, and feeds the narrative that the system is rigged, corrupted, broken, more like gambling instead of investing, etc. The little guy, retiree, retail investor is thrown into the swamp with big fish, HFT, day traders, speculators and hedge funds. They can't compete or keep up, but their money is still swimming in the swamp. Billions get taken out and put into Treasurys or bonds, as flight to safety, turns into earning losses (can't even outrun inflation). Even parking cash in a bank doesn't yield enough interest to keep up. Everyone knows it's a crappy way to save or invest, but there aren't many alternatives. Especially for older geezers like me, who don't have a 40-50 long term time frame like you.
Huh? No, I said the opposite. With the added observation that large financial investment firms (and big banks with trading desks) have hired math and physics quants to exploit technology + time, seeking profit at the end of the Long Tail, in nano-seconds and micro-dollars. Sometimes it's an even simpler failure, by giving computers the benefit of the doubt, and losing track of money (MF Global) or where it goes (Standard Charter).Quote:
Originally Posted by wiggin
If it's trading (not investing) and regulators can't keep up (with speed of innovation) and retail investors are losing confidence....then separate and/or different exchanges could be an alternative. Never said it was a panacea but a suggestion, a different type of innovation.Quote:
No one ever said it was investing, we all agree it's trading. You're straying from the question of separate exchanges into a broader complaint about regulation. There may be some worthwhile arguments in there, but they're irrelevant to your suggestion about making two separate exchanges.
True, the long-term retail investor is interested in purchase prices (plus P/E ratios, maybe dividend payments, etc), but also growth and value when selling. In the short-term, day traders, speculators, hedge funds and HFT can impact those buy-and-sell prices. They can cloud fundamentals, trends, and decisions. (Notice the wired example of buying Yahoo?) Even the term "long-term" has a new meaning! That's no longer something to be ignored, but forces both the financial advisor AND the little guy into trying to "time" the market. Weeding out errant highs or lows, bubbles, volatility, dead-cat bounces, irrational rallies, etc. That's not a healthy stock market.Quote:
You have yet to prove that high frequency trading or algorithmic trading significant changes the long term price of an equity, and that's all that matters to the retail investor. This isn't a question of 'competing' - the losers to HFTs are day traders, not the long term investor. The long term investor gets significant benefits and likely fairly little in the way of costs. You have yet to address the very big problems I raised with having separate exchanges and how it could possibly benefit a retail investor.
Local or regional exchanges have hurdles to clear, that's true too. It's why it's taken several years for F & M and LanX to be up and running. But its appeal is similar to any other local, small, or slow-growth movement. Whether from the business or investor side, plenty of people don't want to be dragged into the big swamp, the big boards, where all the big fish compete to swallow the little fish.....and muck things up. Adding choices and options are good things.
That's not to say people won't still buy blue chips or conglomerates, diversify holdings by industry sectors, or re-calibrate over time. Instead, I see it more like a broader diversification -- that doesn't force the investor to have all eggs in one basket, when that "basket" means....the NYSE or Nasdaq. For small business, it'd be an option to get investor capital without having to use a big bank for a loan. For the growing business, it'd be an option to making the big shift into publicly held shares on big board IPOs, while retaining some self-determination.
Hope that made my pov clearer (?)
At the risk of making things more complicated....."anyone in the market" has a pretty long half-life, thanks to our tax structure and inherited wealth. That has to be largely ignored by new investors, because they have no choice but to buy-hold-sell based on today's valuations. Unfortunately, newcomers don't have the luxury of backward looking valuations, before the NYSE changed, before commodities were traded on separate mercantile or grain exchanges....and before carry-trades could make gains look like losses (and turn profits into zero taxation asset vehicles).Quote:
Originally Posted by LittleFuzzy
For anyone who's been following US political debate about taxes, tax shelters, or tax 'evasion'...or how Romney could possibly have million$ in an IRA (despite annual contribution caps) that's the answer: he inherited that IRA wealth from his parents.
Not actually so complicated because it doesn't actually matter. Any effects are so small as to be near or below the threshold of detection, much less significance and can be ignored as meaningless for your "retail-level investor."
My post was saying the exact same thing Wiggin said in his last paragraph
Quote:
You have yet to prove that high frequency trading or algorithmic trading significant changes the long term price of an equity, and that's all that matters to the retail investor. This isn't a question of 'competing' - the losers to HFTs are day traders, not the long term investor. The long term investor gets significant benefits and likely fairly little in the way of costs.
Taxes matter when making investment decisions (hence, the argument about taxing capital gains). Hell, when you die even matters from a death and estate tax standpoint.
I already addressed things that affect prices "long-term" -- including a different (shorter) definition of LT, timing buy-sell points, speed/volatility that comes from e-trades, HFT, and volume, etc. Plus bubbles. People who bought during the dotcom bubble can't "hold" and still see profit. People who sold during the "meltdown" -- including retirees whose IRAs timed-out during that time period -- lost quite a bit.
We've argued before about the role of "speculators" and day traders in markets too, especially oil and commodities. One side says they benefit end prices and consumers, the other side says they add too much volatility and anxiety that hurts prices/consumers. Every variable matters in some way, that's all I'm saying. Not just for equities, and not just from HFT. Glass half-full or half-empty, just don't dismiss all contents in the glass.
Not in any way that makes HFTs inimical to your "retail investor."Quote:
Originally Posted by GGT
Every variable has an effect. That's not the same thing. Not at all. For it to matter the effect has to have a significant impact, it has to have a fair bit of heft to itQuote:
We've argued before about the role of "speculators" and day traders in markets too, especially oil and commodities. One side says they benefit end prices and consumers, the other side says they add too much volatility and anxiety that hurts prices/consumers. Every variable matters in some way, that's all I'm saying. Not just for equities, and not just from HFT. Glass half-full or half-empty, just don't dismiss all contents in the glass.
HFT has only been in use, and in growing numbers, for about five years. You're asking me to 'prove' something that can't/won't be proven for another few years. Meanwhile, you're using the short term to assert the claim that it has no impact on today's retail investor... :think:
In the bigger picture, since the software programmers, quality control folks, risk analysts, trading desks, brokers/agents, and regulators all say it matters -- I figure it matters. No idea what the heft will be, it's an unknown unknown. Just like the synthetic derivatives and CDOs of the sub-prime crisis and meltdown seemed like a good idea to some, WMD to others. I'm fine with being a skeptic. ;)
Long-term or short-term?Quote:
Originally Posted by GGT
Nifty chart in motion :eek:
http://www.nanex.net/aqck/2804.HTML