I'm not sure if there should be any outrage over this.
It was already known that they released the data 5 minutes early to subscribers who wished to pay more. It's not really a stretch to think they'd offer tiers for people who wished to pay more.
The data is, after all, legally theirs to release, and Michigan University knew and approved it.
If this is something people don't like then they can vote with their feet and move to a different news source.
Yes, unsurprising considering how much traders pay to get their servers co-located in exchanges' data centres. This is orders of magnitude more time. But should they have been more open about it?
There is nothing inherently wrong with choosing to release reports that you own or have exclusive license to in any way you want, tiered or not. I pay for Showtime, and therefore I get to see new Homeland episodes before people downloading from free torrent sites get to. I have chosen to pay; they have chosen not to pay. Same thing here.
A significant chunk of the money Reuters pays for the reports, which they recoup through subscriptions, winds up going to pay for the production of the reports anyway. The only thing that surprises me about this is how much money they are likely leaving on the table. They should let people bid for early access.
With the idea being that those who bought access had an unfair advantage in trading with the general public.
However, if "everyone" knows that some companies can buy these reports minutes ahead of time, then when people lose out on trades in the minutes afterwards, isn't it the fault of the people taking the trade?
It's not wise to trade with people who might have more knowledge than you. It seems like there's a reasonable expectation that the people on the other sides of these trades should know about the paid access.
There is nothing illegal, immoral, or unethical about collecting information or paying someone to collect information to give you an advantage in a trade. Nearly every trade that happens has some information assymetry associated with it. It is not a secret in any way that there are tiers of access to reuters data. We could enact some sort of law that makes this illegal, but that seems heavy handed and unlikely to have a positive impact on the market.
We have decided that it should be illegal for public companies to give information selectively for the purposes of trading. We could easily change this law, but history has shown us that it isn't heavy handed and does have a positive impact on the market.
Giving the information more quickly to one group rather than the others is a form of lightning-fast pump and dump.
Also, the sentiment numbers are rigged. The one phrase you can utter that will get your mike cut on CNN is: "The fix is in." Specifying a little abnormal bump up or down in the indicator, for a price, can be a very lucrative business, so long as you don't get caught.
Why don't we just have turn based trading with turns of 1 or 24 hours and then have a blackout period near turn ends when relevant information should not be divulged?
I don't understand. So you can make side bets in real time with others who choose to do so. The point is that those who choose not to do so don't have to compete on millisecond scale response times to news postings.
Maybe, but I suspect you'd end up with worse prices.
It would be hard to enforce a true "blackout period". Traders trade on all kinds of information, not just controlled data. Wouldn't liquidity tend to be lower on a non-realtime exchange, because everyone is essentially trading on outdated information, resulting in worse prices than the realtime exchange?
But I know very little about this stuff, so maybe I'm completely wrong.
The blackout scheme is a very secondary thing. If you make interval 1 hour, then the rapid response advantage only applies to data made available in the seconds before the trading event. Rather than having a millisecond advantage on every trade, the fat pipe guys have a millisecond advantage on every 3.6 millionth trade.
I think prices would be better for people not playing the low latency response game. Liquidity within an hour would probably cost a premium as you'd have to get the money from someone on the real time market.
There is no such thing as "the market" to time market opens and closes around. For instance the S&P futures contract which is impacted by this is only closed from 4pm-5pm. You can trade equities on several platforms during this time.
That actually would be really interesting, because you're essentially creating an auction on the value of information. It does you no good to get the news a nanosecond after your competitor, because in these markets (with large amounts of leverage available) they've already bought up all the affected securities and arbitraged out the difference. So it behooves you to bid up to your expected profit from having that information, in an attempt to get the information first. The transaction price gives you the dollar value of the information to the highest bidder.
Right, which is why major hedge funds heavily optimize their code, use C instead of Java to avoid GC delays, and colocate their boxes on the exchange to minimize speed-of-light delays. We already know the premium for executing trades faster via solid IT: it's a lot.
You'll notice the contract specifies a +/- 500ms slack time which would be a limiting factor in providing the data to the clients on that specific of a timing bound. Realtime constraints that are that specific can be really hard to do in bulk.
You could easily group things into 500ms classes though. But 500ms is kind of an eternity in these types of things and you'll never really have someone thinking that they have a system fast enough to beat someone who started 500ms earlier.
500ms is a long time for an HFT algorithm, from what I understand. If my algo takes the Reuters feed as input (which I imagine to be a good leading indicator for trading) and I have a 500ms lead, it could be the one that bought the stock right before yours did (and enough shares to drive up the price adequately to slurp up all the margin).
2 seconds can be a lifetime in this business. At least in Sweden one of the stock exchange offers server space in the same machine hall as the exchange. To minimize the physical distance.
Yes, this gives "elite" traders an advantage over the only-somewhat-elite traders, but I would suspect that most sane traders who aren't doing hyper-HFT-type trading will be unaffected. So it's a meh for me.
'[Reuters] explained that this disclosure can be found on the "Machine Readable News" product page of its Website, under a drop-down menu for "suite components." ' - CNBC
(It's actually worse - it's six clicks from the home page to the document.)
' "...I eventually had to go down to the cellar to find [the plans]."
"That's the display department."
"With a torch."
"Ah, well, the lights had probably gone."
"So had the stairs."
"But look, you found the notice, didn't you?"
"Yes," said Arthur, "yes I did. It was on display in the bottom of a locked filing cabinet stuck in a disused lavatory with a sign on the door saying 'Beware of The Leopard'." ' - Douglas Adams
There is a certain amount of overhead involved in the creation of the dataset. Hiring people to call and collect answers, filtering the data, formatting it, fielding questions about it, etc. The University of Michigan seems to partially pay for this by the fees they charge Reuters. I am not sure if they make a profit on it or, if so, how much. But you could probably found some company to collect and provide the same amount and quality of data for cheaper.
However, even if you were able to beat them on cost, you wouldn't be able to charge nearly the same amount that UofM does for it. The reason is that the UofM data goes back for over three decades, so traders can do all sorts of backtesting with it. They are also a recognized "brand" in this field and over time sources of trading information like the UofM report become ingrained in traders' minds as "indicators" of one type or another. So, for various practical and psychological reasons, your hypothetical startup would have to charge a much lower price for the same data and spend many years earning the confidence of traders.
Ironically, you might be able to make more money by selling the data as an early predictor of the UofM report. Since traders know that the UofM report can be a market moving event, any early predictor of it (even one with ~90% accuracy) would be valuable. This would probably just force UofM to collect and release the data earlier and with less polish though, negating much of your advantage.
Yeah, the mind-bending part of the whole thing is that the reason the traders are making trades based on the information ISN'T because the data predicts the long term (or even short term) profitability of any company; they are making trades based on it because they know the number will affect the STOCK PRICE of the company... at least temporarily.
It doesn't matter how good your own data is if the data's release doesn't affect stock prices.
Your last point is kind of what I had in mind with my question. Not to try to compete with UofM but to gather data to predict their results and use it for your own trades. You could hold the results private and play the market to beat even the high-frequency guys. You could even do a shoddy job of collecting the data to keep costs down but maintain, say, a 90% confidence in your prediction of UofM's numbers, that's enough to beat the market 9 out of 10 times.
You don't factor in the number of times UofM 'gets it wrong'. Even given their influence you wouldn't be winning 9 out of 10 times. But it could still be an edge.
I fail to see the problem. If you don't want to have any 'tiers' in third party analysis info, logically it follows that all reports should be published free of charge.
I don't see the issue here. Now that this is common knowledge, if you don't subscribe to the Thomson Reuters feed, DO NOT TRADE until 10am and the price has settled. Why would you ever participate in a transaction with someone who you know has more information than you, and you know this information asymmetry will be equalized within 5 minutes?
Someone at Michigan should make a startup that builds a system to sell this information at market value, which most certainly is way more than $1 million per year.
42 comments
[ 4.2 ms ] story [ 96.8 ms ] threadIt was already known that they released the data 5 minutes early to subscribers who wished to pay more. It's not really a stretch to think they'd offer tiers for people who wished to pay more.
The data is, after all, legally theirs to release, and Michigan University knew and approved it.
If this is something people don't like then they can vote with their feet and move to a different news source.
Once again Nanex is all over this.
A significant chunk of the money Reuters pays for the reports, which they recoup through subscriptions, winds up going to pay for the production of the reports anyway. The only thing that surprises me about this is how much money they are likely leaving on the table. They should let people bid for early access.
However, if "everyone" knows that some companies can buy these reports minutes ahead of time, then when people lose out on trades in the minutes afterwards, isn't it the fault of the people taking the trade?
It's not wise to trade with people who might have more knowledge than you. It seems like there's a reasonable expectation that the people on the other sides of these trades should know about the paid access.
We have decided that it should be illegal for public companies to give information selectively for the purposes of trading. We could easily change this law, but history has shown us that it isn't heavy handed and does have a positive impact on the market.
Also, the sentiment numbers are rigged. The one phrase you can utter that will get your mike cut on CNN is: "The fix is in." Specifying a little abnormal bump up or down in the indicator, for a price, can be a very lucrative business, so long as you don't get caught.
It would be hard to enforce a true "blackout period". Traders trade on all kinds of information, not just controlled data. Wouldn't liquidity tend to be lower on a non-realtime exchange, because everyone is essentially trading on outdated information, resulting in worse prices than the realtime exchange?
But I know very little about this stuff, so maybe I'm completely wrong.
I think prices would be better for people not playing the low latency response game. Liquidity within an hour would probably cost a premium as you'd have to get the money from someone on the real time market.
I'm not sure if I'm joking or not.
You could easily group things into 500ms classes though. But 500ms is kind of an eternity in these types of things and you'll never really have someone thinking that they have a system fast enough to beat someone who started 500ms earlier.
That said, serious traders who object to this tax can predict the signal from more-primary sources.
(It's actually worse - it's six clicks from the home page to the document.)
' "...I eventually had to go down to the cellar to find [the plans]."
"That's the display department."
"With a torch."
"Ah, well, the lights had probably gone."
"So had the stairs."
"But look, you found the notice, didn't you?"
"Yes," said Arthur, "yes I did. It was on display in the bottom of a locked filing cabinet stuck in a disused lavatory with a sign on the door saying 'Beware of The Leopard'." ' - Douglas Adams
There is a certain amount of overhead involved in the creation of the dataset. Hiring people to call and collect answers, filtering the data, formatting it, fielding questions about it, etc. The University of Michigan seems to partially pay for this by the fees they charge Reuters. I am not sure if they make a profit on it or, if so, how much. But you could probably found some company to collect and provide the same amount and quality of data for cheaper.
However, even if you were able to beat them on cost, you wouldn't be able to charge nearly the same amount that UofM does for it. The reason is that the UofM data goes back for over three decades, so traders can do all sorts of backtesting with it. They are also a recognized "brand" in this field and over time sources of trading information like the UofM report become ingrained in traders' minds as "indicators" of one type or another. So, for various practical and psychological reasons, your hypothetical startup would have to charge a much lower price for the same data and spend many years earning the confidence of traders.
Ironically, you might be able to make more money by selling the data as an early predictor of the UofM report. Since traders know that the UofM report can be a market moving event, any early predictor of it (even one with ~90% accuracy) would be valuable. This would probably just force UofM to collect and release the data earlier and with less polish though, negating much of your advantage.
It doesn't matter how good your own data is if the data's release doesn't affect stock prices.
That would make no sense at all.
Someone at Michigan should make a startup that builds a system to sell this information at market value, which most certainly is way more than $1 million per year.