So... penny per trade tax... that works this out to what 17 million a day?
The day that such a tax is enabled, all of the professional traders will simply account for it, the way they already account for clearing fees, exchange fees, and SEC "assessments" (roughly equivalent to a transaction tax). You won't see $17M on the first day, you will see some smaller number, along with lower trading volume because of the tighter profit margins. It's a Laffer curve, so nobody knows how much lower... but everyone who understands the system knows that calculating future revenue of a transaction tax from current trading volume is a lie.
If for example automated trades dont go the way wall-street firms like, they can have them rolled back.
I keep hearing this, but nobody seems willing to provide the evidence. Can you cite a recent example? The "flash crash" on May 6, 2010 prompted the standardization of trade bust rules, which AFAIK shut down any discretionary-bust shenanigans.
Market Making, or as we know it to be pump and dump, is a real algorythm employed daily by firms.
You are wholly incorrect on this. Market making is the practice of putting two-sided quotes into the market to provide liquidity for others to trade against, with the incentive of capturing the spread when volatility is low. Market makers make the most money when prices oscillate around a small range. Pump and dump is the practice of publishing false information with the expectation that it will materially move the market price, but only temporarily. Pump-and-dumpers want to capture one single big price movement, and they are likely to be liquidity takers.
somehow right if the exchange is the middle-man in the trade
The exchanges publish a stream of information about quotes and trades, in realtime. They don't delay trade printing by up to 30 seconds. They show limit orders in the book, which ensures a fair playing field in the matching engine. They don't allow one participant to "turn off" matching with another participant. They publish APIs, fees, and rulebooks on their web sites. For all of these reasons, I prefer to have the exchange as the middle man.
I can't speak for Mitt, but what I want is to get rid of these bloated exchanges and replace them with cut-throat competitors.
That's what BATS and DirectEdge have been trying to do, for several years now. It's hard to de-throne NYSE and NASDAQ because of the network effect: traders go to NYSE/NASDAQ markets because there is liquidity there. Liquidity is created by traders. So, if everyone moved tomorrow the old exchanges would die quite fast, but that's similar to the thermodynamic probability of all of the particles in your body moving in the same direction at once.
Perhaps you shouldn't project your own shortcomings onto others...
Do you plan on becoming less wise as you age? If not, then you will at some point probably realize mistakes you have made in your youth. This isn't a knock against the younger generations, it is the natural outcome of learning.
How much of that worth is available to be put into a high-risk investment? Suppose we are generous and say that some 65-year old is willing to risk 1/4 of his worth for the chance to earn more than the going interest rate. That's $50,000. Now suppose that in order to diversify risk, that gets split up into 5-10 different stocks or ETFs. Now we're down to $5-10k apiece. A $5 trading fee on that amount is 10-20 basis points, which is lower than changing currency but still two orders of magnitude above the rate you quoted. And this still assumes trading the entire position in one stock/ETF. Suppose that the investor rebalances the whole portfolio once every quarter, and each rebalance only involves a trade in half of the stocks. A $5 per-trade fee works out to $50-100 annually, again 10-20 basis points. That doesn't sound so good to me, and it's certainly nothing close to the notional fees payed by HFT or other professional traders.
HFT routinely places insincere bids and yanks them back before a transaction completes.
That is not the same as the Schroedinger's apple that you described above. Trading on the stock market is atomic: If HFT yanks a bid before anyone else hits it, then no transaction has started and so nothing completes. If HFT tries to yank a bid that has already been hit, then they fail. Their bid has already executed, you can't cancel it once the exchange has matched it with another order.
If they REALLY screw up, because of their size they get a do-over where all transactions in a stock get rolled back.
If they really screw up, then they bleed money on ~150 NYSE-listed stocks for 45 minutes. All trades that occur within the predefined price limits stand, and all trades that occur outside of those price limits are busted because that's what NYSE has decided in advance. There may have been some shady "takesies backsies" in the past. To my knowledge, all of that stopped on the Flash Crash day, when traders realized how scary it was to allow the exchange to roll back trades on a discretionary basis. Knight Capital's debacle is a direct counterexample to your claim. Can you provide any recent examples of discretionary rollbacks in the stock market?
And flash yanks his offer back before it completes and walks away no worse for the wear.
Is this Schroedinger's apple? Either flash bought the apple from Jim, or he didn't. If he has already bought it, he takes on the risk of short-term apple volatility. If he has not already bought it, then Jon is free to buy it from Jim.
Maybe nobody else wants that apple? Jon wants the apple for $0.50, but since he's pissed at flash maybe he only offers to buy it for $0.49 from flash. Since flash doesn't want the apple, he may fold and sell it at a loss to Jon, before another apple seller shows up and pushes the market price down further.
In other words, the only way that Jon is hurt by flash is if he caves in and willingly buys the apple for more than the original asking price.
But you're right about the getting-popped-in-the-nose part. As our society has become de-personalized, more and more actions occur which would not be acceptable face-to-face. This is probably a long-term problem, though it has nothing to do with HFT.
Another thing about HFT: It's existence dramatically limits the average investor (John Q. Public) from profiting on these differences between share price and perceived value: Even if he logs into E-trade the moment he has the idea, the HFTs have had the data for 10 minutes or more and have already profited 10,000 times and killed the average investor's profits.
If John Q. Public's "ideas" are observations that anyone in the public can see, they don't exactly sound like ideas to me. More like, John heard some news, and he wants to trade on it... but oh no! Everyone else in the market heard the same news, and the information is now priced into the market. From his perspective (and to his chagrin), the efficient market hypothesis holds.
Now, if John Q. Public has an idea that really isn't some momentum borne out of news—something that's genuinely novel—then he can trade against all of the liquidity available in the market. And when he's right, he'll make a boatload of cash. But the market makers he trade with won't care, because they'll have long since captured their <$0.01/share profit and moved on.
The above is really sort of a big question. I have no idea what HFT is really doing.
That's what HFT market-makers do (whether registered or de-facto). They compete with each other to quote a spread that is tight enough to receive the order flow from player1/player2, but wide enough so that they are still profitable on average. HFT market makers are good for the market in most conditions. The biggest downside is that they have long since starved out all of the old specialist market-makers, and so when uncertainty arrives in the real world, many of them will either shut off or widen their spread out to avoid getting clobbered by news. Since the specialists are gone, that widening can induce "surprise" volatility in the market.
Dark pools are still required to match your orders within the national bid-ask spread. That is the regulation that matters the most for retail traders, IMHO. Also, note that (except for registered market makers) all trading on the primary exchanges is equally anonymous. You can trade with George Soros on a dark pool, or on the NASDAQ, and you're going to get crushed just the same either way.
What it comes down to is that they cost a lot to support (force investement in systems) but don't always deliver as much value as other more lucrative clients that trade using our algos rather than HFT DMA merry-making in our dark pool.
If you didn't have HFT playing in your dark pool, that would probably mean that you wouldn't have market makers. That would increase the time-to-fill for all of your market-taking clients, as they would have to match up with each other, by happenstance. Some of them would lose interest in waiting, and volume would dry up on your dark pool. Or was there a time when you had a dark pool and no market-makers, but only big block trades going through on rare occasion?
If it's riskier, then in aggregate, the profit will even out.
That's not true. Consider that high-risk investment strategies may appear similar to the Martingale strategy, or a falling knive. These strategies may be profitable on average, in the long run; but the capital requirements to avoid getting "stopped out" are prohibitive for small-capitalization investors. Such strategies may have a higher expected value, but may be unfeasible to attempt due to the risk. For example, VC investors pour a lot of money into a variety of new companies which are all probably going to fail, because the expected returns on a single success are so high. The high risk keeps many investors away from VC, but clearly the expected reward is higher in aggregate for the VC investors, because that is what they choose to do.
BTW, Plasma TV is old technology. I haven't seen a new one sold in years.
I know that this is getting off on a tangent, but I can't let it go: plasma TV technology is continuing to improve alongside LCD. The model that I purchased recently (Panasonic VT series) had some clear advantages:
A wider dynamic range than any of the competing LCDs: good for watching movies in a darkened room.
Lower input latency than any of the competing LCDs: good for twitch-games like first-person shooters or Street-Fighter-style fighting games. Note that this information is typically not published by manufacturers (it really only matters for games where reaction time is paramount), but there are some dedicated individuals who have done their own timing tests.
Native support for 24fps source material, i.e. no frame interpolation. Not specifically an advantage over high-end LCDs, but it is a recent feature in the "history of flat-panel TVs".
Comparable motion-estimation algorithms to reduce judder on material with large pans, such as the BBC Planet Earth "series".
More screen per dollar spent than competing LCDs.
If memory serves, one of the top contenders from the LCD side was the Sony D8000.
I've only done work in the field in this century—but the last time I had to worry about CPU architecture was probably 6 months ago, when my employer upgraded the chipset for some high-performance internal software. The importance of computer architecture to an individual developer simply varies based on workload: it matters for mine, it doesn't matter for yours, big whoop. However, the importance of computer architecture to the field as a whole should not be underestimated.
The implication is that they sell the things for money and then get a replacement by claiming it is lost
Yes, I understand the implication there, but it's not exactly damning evidence of large-scale fraud; just a surprising statistic worth further investigation. That press release was over a year ago; has the state followed through and discovered evidence of fraud? Has anyone had their EBT access/re-issue revoked, payed a fine, or been locked up? If cards were sold, then surely the buyer got some value–i.e. purchased goods/food–from the card. And surely the seller was able to recoup those costs by reporting cards as stolen and receiving credit for the recently-spent values. Someone with access to the data could do the math, and then report "the state of Ohio covered $Y of goods purchased on reported-stolen cards, including $Z from person X, who reported a stolen card 75 times in 2011." I don't have that data.
The State of Ohio reissued nearly 340,000 food stamp EBT cards in 2011 – and 17,000 recipients received 10 or more reissued cards
Maybe I'm reading that article wrong, but it sounds like 5% of the EBT users lost their cards ten times in the year. Wow, that sounds pretty bad, but it could simply be the set of people who would lose their heads if they weren't firmly attached to necks. The article seems to say nothing about (1) any cards being used after being reported as lost/stolen, or (2) multiple cards being used concurrently by a single person. So yeah, there's a waste of money in reissuing cards, but I don't see any fraud specifically highlighted. Just the lack of analysis into potential fraud.
Ok, so let's assume that the person who reported 75 lost/stolen EBT cards did so immediately after a large purchase, so that the cost of the large purchase would be credited back to the account. Now, that would be pretty clear fraud. It also has a price tag: total value of EBT purchases made after a card was claimed to have been "stolen", but before it could be deactivated. Where is that number?
Capital gains (how the rich make most of their income) should be treated no differently than regular income.
I hear this a lot, but I think it is a knee-jerk reaction to hearing about rich people paying such low capital gains taxes. The reason that capital gains taxes are lower is because most activities that generate such gains are inherently riskier than working for a paycheck. So I think it does make sense to start capital gains taxes at a low rate, to encourage small-time investing. The problem, as I see it, is that for people with lots of capital to throw around, it is much easier to diversify the risks. At this point, the lower tax rate is not actually creating an incentive to invest, because the risk of catastrophic loss is comparatively very low. Short-term capital gains are already taxed as income. Long-term capital gains tax rates should probably be progressive enough to eliminate the "perverted" tax incentive on the Warren Buffetts of the world.
Because taxes are levied on capital gains (i.e. only when selling his shares) and not on capital itself, then taxing Steve would have done about nothing. He carried many of them to his grave. Though I agree with you about what they represented to him: control. Stock ownership is only about profit when you sell your stake. Holding your stake is holding control.
The low level details of the CPU implementation are a very small corner of computer science.
Your own quotation doesn't seem to support your assertion:
The IEEE Computer Society identifies four areas that it considers crucial to the discipline of computer science: theory of computation, algorithms and data structures, programming methodology and languages, and computer elements and architecture.
Emphasis mine. One of the four "crucial areas" is the CPU implementation.
Slashdot Mirror
The day that such a tax is enabled, all of the professional traders will simply account for it, the way they already account for clearing fees, exchange fees, and SEC "assessments" (roughly equivalent to a transaction tax). You won't see $17M on the first day, you will see some smaller number, along with lower trading volume because of the tighter profit margins. It's a Laffer curve, so nobody knows how much lower... but everyone who understands the system knows that calculating future revenue of a transaction tax from current trading volume is a lie.
I keep hearing this, but nobody seems willing to provide the evidence. Can you cite a recent example? The "flash crash" on May 6, 2010 prompted the standardization of trade bust rules, which AFAIK shut down any discretionary-bust shenanigans.
You are wholly incorrect on this. Market making is the practice of putting two-sided quotes into the market to provide liquidity for others to trade against, with the incentive of capturing the spread when volatility is low. Market makers make the most money when prices oscillate around a small range. Pump and dump is the practice of publishing false information with the expectation that it will materially move the market price, but only temporarily. Pump-and-dumpers want to capture one single big price movement, and they are likely to be liquidity takers.
The exchanges publish a stream of information about quotes and trades, in realtime. They don't delay trade printing by up to 30 seconds. They show limit orders in the book, which ensures a fair playing field in the matching engine. They don't allow one participant to "turn off" matching with another participant. They publish APIs, fees, and rulebooks on their web sites. For all of these reasons, I prefer to have the exchange as the middle man.
That's what BATS and DirectEdge have been trying to do, for several years now. It's hard to de-throne NYSE and NASDAQ because of the network effect: traders go to NYSE/NASDAQ markets because there is liquidity there. Liquidity is created by traders. So, if everyone moved tomorrow the old exchanges would die quite fast, but that's similar to the thermodynamic probability of all of the particles in your body moving in the same direction at once.
You may as well omit that long-ass conditional and just say "the simplest way to accomplish something."
Do you plan on becoming less wise as you age? If not, then you will at some point probably realize mistakes you have made in your youth. This isn't a knock against the younger generations, it is the natural outcome of learning.
How much of that worth is available to be put into a high-risk investment? Suppose we are generous and say that some 65-year old is willing to risk 1/4 of his worth for the chance to earn more than the going interest rate. That's $50,000. Now suppose that in order to diversify risk, that gets split up into 5-10 different stocks or ETFs. Now we're down to $5-10k apiece. A $5 trading fee on that amount is 10-20 basis points, which is lower than changing currency but still two orders of magnitude above the rate you quoted. And this still assumes trading the entire position in one stock/ETF. Suppose that the investor rebalances the whole portfolio once every quarter, and each rebalance only involves a trade in half of the stocks. A $5 per-trade fee works out to $50-100 annually, again 10-20 basis points. That doesn't sound so good to me, and it's certainly nothing close to the notional fees payed by HFT or other professional traders.
Ball mice, with lots of hair! Give them hell.
Oh yeah, that explains why the Chicago Stock Exchange is totally dominating the exchange space.
That is not the same as the Schroedinger's apple that you described above. Trading on the stock market is atomic: If HFT yanks a bid before anyone else hits it, then no transaction has started and so nothing completes. If HFT tries to yank a bid that has already been hit, then they fail. Their bid has already executed, you can't cancel it once the exchange has matched it with another order.
If they really screw up, then they bleed money on ~150 NYSE-listed stocks for 45 minutes. All trades that occur within the predefined price limits stand, and all trades that occur outside of those price limits are busted because that's what NYSE has decided in advance. There may have been some shady "takesies backsies" in the past. To my knowledge, all of that stopped on the Flash Crash day, when traders realized how scary it was to allow the exchange to roll back trades on a discretionary basis. Knight Capital's debacle is a direct counterexample to your claim. Can you provide any recent examples of discretionary rollbacks in the stock market?
Is this Schroedinger's apple? Either flash bought the apple from Jim, or he didn't. If he has already bought it, he takes on the risk of short-term apple volatility. If he has not already bought it, then Jon is free to buy it from Jim.
Maybe nobody else wants that apple? Jon wants the apple for $0.50, but since he's pissed at flash maybe he only offers to buy it for $0.49 from flash. Since flash doesn't want the apple, he may fold and sell it at a loss to Jon, before another apple seller shows up and pushes the market price down further.
In other words, the only way that Jon is hurt by flash is if he caves in and willingly buys the apple for more than the original asking price.
But you're right about the getting-popped-in-the-nose part. As our society has become de-personalized, more and more actions occur which would not be acceptable face-to-face. This is probably a long-term problem, though it has nothing to do with HFT.
If John Q. Public's "ideas" are observations that anyone in the public can see, they don't exactly sound like ideas to me. More like, John heard some news, and he wants to trade on it... but oh no! Everyone else in the market heard the same news, and the information is now priced into the market. From his perspective (and to his chagrin), the efficient market hypothesis holds.
Now, if John Q. Public has an idea that really isn't some momentum borne out of news—something that's genuinely novel—then he can trade against all of the liquidity available in the market. And when he's right, he'll make a boatload of cash. But the market makers he trade with won't care, because they'll have long since captured their <$0.01/share profit and moved on.
That's what HFT market-makers do (whether registered or de-facto). They compete with each other to quote a spread that is tight enough to receive the order flow from player1/player2, but wide enough so that they are still profitable on average. HFT market makers are good for the market in most conditions. The biggest downside is that they have long since starved out all of the old specialist market-makers, and so when uncertainty arrives in the real world, many of them will either shut off or widen their spread out to avoid getting clobbered by news. Since the specialists are gone, that widening can induce "surprise" volatility in the market.
Dark pools are still required to match your orders within the national bid-ask spread. That is the regulation that matters the most for retail traders, IMHO. Also, note that (except for registered market makers) all trading on the primary exchanges is equally anonymous. You can trade with George Soros on a dark pool, or on the NASDAQ, and you're going to get crushed just the same either way.
If you didn't have HFT playing in your dark pool, that would probably mean that you wouldn't have market makers. That would increase the time-to-fill for all of your market-taking clients, as they would have to match up with each other, by happenstance. Some of them would lose interest in waiting, and volume would dry up on your dark pool. Or was there a time when you had a dark pool and no market-makers, but only big block trades going through on rare occasion?
That's not true. Consider that high-risk investment strategies may appear similar to the Martingale strategy, or a falling knive. These strategies may be profitable on average, in the long run; but the capital requirements to avoid getting "stopped out" are prohibitive for small-capitalization investors. Such strategies may have a higher expected value, but may be unfeasible to attempt due to the risk. For example, VC investors pour a lot of money into a variety of new companies which are all probably going to fail, because the expected returns on a single success are so high. The high risk keeps many investors away from VC, but clearly the expected reward is higher in aggregate for the VC investors, because that is what they choose to do.
I know that this is getting off on a tangent, but I can't let it go: plasma TV technology is continuing to improve alongside LCD. The model that I purchased recently (Panasonic VT series) had some clear advantages:
If memory serves, one of the top contenders from the LCD side was the Sony D8000.
I've only done work in the field in this century—but the last time I had to worry about CPU architecture was probably 6 months ago, when my employer upgraded the chipset for some high-performance internal software. The importance of computer architecture to an individual developer simply varies based on workload: it matters for mine, it doesn't matter for yours, big whoop. However, the importance of computer architecture to the field as a whole should not be underestimated.
Yes, I understand the implication there, but it's not exactly damning evidence of large-scale fraud; just a surprising statistic worth further investigation. That press release was over a year ago; has the state followed through and discovered evidence of fraud? Has anyone had their EBT access/re-issue revoked, payed a fine, or been locked up? If cards were sold, then surely the buyer got some value–i.e. purchased goods/food–from the card. And surely the seller was able to recoup those costs by reporting cards as stolen and receiving credit for the recently-spent values. Someone with access to the data could do the math, and then report "the state of Ohio covered $Y of goods purchased on reported-stolen cards, including $Z from person X, who reported a stolen card 75 times in 2011." I don't have that data.
Scratch that... there was a Dec 2013 fraud conviction. And the USDA estimates that Ohio may be responsible for $30 million in food-stamp fraud (residents or business owners, not the stat itself). So there is documented fraud... but not from the EBT cardholders themselves.
Maybe I'm reading that article wrong, but it sounds like 5% of the EBT users lost their cards ten times in the year. Wow, that sounds pretty bad, but it could simply be the set of people who would lose their heads if they weren't firmly attached to necks. The article seems to say nothing about (1) any cards being used after being reported as lost/stolen, or (2) multiple cards being used concurrently by a single person. So yeah, there's a waste of money in reissuing cards, but I don't see any fraud specifically highlighted. Just the lack of analysis into potential fraud.
Ok, so let's assume that the person who reported 75 lost/stolen EBT cards did so immediately after a large purchase, so that the cost of the large purchase would be credited back to the account. Now, that would be pretty clear fraud. It also has a price tag: total value of EBT purchases made after a card was claimed to have been "stolen", but before it could be deactivated. Where is that number?
Be vigilant, my friend. The canadian storm is coming.
I hear this a lot, but I think it is a knee-jerk reaction to hearing about rich people paying such low capital gains taxes. The reason that capital gains taxes are lower is because most activities that generate such gains are inherently riskier than working for a paycheck. So I think it does make sense to start capital gains taxes at a low rate, to encourage small-time investing. The problem, as I see it, is that for people with lots of capital to throw around, it is much easier to diversify the risks. At this point, the lower tax rate is not actually creating an incentive to invest, because the risk of catastrophic loss is comparatively very low. Short-term capital gains are already taxed as income. Long-term capital gains tax rates should probably be progressive enough to eliminate the "perverted" tax incentive on the Warren Buffetts of the world.
Because taxes are levied on capital gains (i.e. only when selling his shares) and not on capital itself, then taxing Steve would have done about nothing. He carried many of them to his grave. Though I agree with you about what they represented to him: control. Stock ownership is only about profit when you sell your stake. Holding your stake is holding control.
Your own quotation doesn't seem to support your assertion:
Emphasis mine. One of the four "crucial areas" is the CPU implementation.