Bloomberg Admits Terminal Snooping
nytimes.com
nytimes.com
Pretty much every single sophisticated SaaS company in the world collects this data (login frequency, etc.) - it's invaluable for sales, support, debugging, product management, etc. hence it also makes sense that the information is widely available within the firm.
There's obviously an ethical issue in that Bloomberg's reporters shouldn't have had access to that data, but it's worth noting that the information is similar to what everyone else in the SaaS space is collecting.
While I think the green dot/red dot controversy is overblown, the paranoia is driven by the ubiquitous aspect of the terminals combined with the massive number of workflows that are centralized on Bloomberg terminals. Here's a sampling:
* Proprietary email and IM systems integral to interactions with customers and counterparties * Trading systems in virtually every asset class. TOMS, the fixed income system, is the defacto standard for bonds. * One of the top financial news gathering organizations * Best-in-class financial data and analytics
So when you think of what someone could do with access to the system, it almost boggles the mind. Ability to know who is looking at what. Who is saying what to whom. And in certain asset classes, the best possible data on who is buying what. A rogue support person, for example, could do incredible damage to the company. Hopefully they have good mechanisms to control that - the fact that we haven't heard about it thus far is a good sign.
Given the overlap of workflows available on the terminals, plus the relatively low-cost of extra features once you're already paying for the terminal, makes me view Bloomberg as an anti-trust case waiting to happen. I wonder if they think so too - and to feed more paranoia, I've wondered whether that explains why Mayor Bloomberg is in politics. (Either that and/or to promote BGOV - growth in the government sector as wall street retrenches).
They're basically leased windows computers with a dedicated application that displays realtime market data. They run about $20k/yr. They come with a funky keyboard, and the app has all sorts of messaging and searching functions into a wide variety of markets. Typically, the average user uses only a tiny fraction of the capability they have via these terminals (the pork belly futures trader doesn't ever bother with high tech semiconductor trades). Because everybody is on a Bloomberg terminal, and it has a dedicated messaging system, you need a Bloomberg terminal to chat with other traders to make deals - hence the network effect. It also marks you as a "serious" player, even small offices with a couple guys will have these terminals.
Since Bloomberg has total vertical integration, from data source to keyboard, it means they collect, or have the potential to collection incredible amounts of information about their user community. Every chat, every query, every trade, every login, logoff, every keypress.
So the unsurprising question is, why wouldn't other parts of Bloomberg have access to some/all of this data? And even if everybody was terribly offended by that, what are they going to do? They're pretty much locked in at this point.
edit relevant WP http://en.wikipedia.org/wiki/Bloomberg_Terminal
Bloomberg are playing with fire here, partly because they're almost a monopoly, and partly because they could be so easily replaced on the quotes side (if not on the financial news market side). Technology has moved on and exchanges are more open with data than they used to be, and this is probably only going to continue - I'm sure they'd love to cut Bloomberg out of the picture and charge their own fees separately.
There is no justification for Bloomberg journalists to be poking into private account information about Bloomberg clients - the potential for abuse of this position is huge, apart from anything else the journalists could use it for insider trading, which is of course illegal. They could also take information from one trader and sell it to a competitor, there is huge scope for abuse, and the only way to avoid it is to have very clear (and tight) boundaries for the usage of stats collected.
In the same way that newspapers have a firewall between editorial and advertisers, to try to stop advertisers influencing editorial, Bloomberg should have a firewall between the division which sells financial information and that which investigates financial news in order to get that information. I find it astonishing that so many people are defending this sort of privacy invasion as standard practice. It's not acceptable to use information such as queries and logins to measure specific individual actions and find out (for example) how active a particular trader has been in a certain market, then spread that information. It's fine to keep these stats and anonymise them in order to generate statistics about usage, but quite another thing to use them to track individuals and trading patterns. If Google for example used the content of private emails from a CEO to target companies to sell adverts to, there would be outrage, and rightly so.
Insider trading is a risk, but no-more so than with everyone else in the financial industry. If you work in finance typically there's pretty tight restrictions on what you're allowed to trade on a personal account (i.e. no shorting, no holding equities for less than 30 days, etc.) and you have to register any non-blind investments (both so things like insider trading can be tracked but also to prevent conflicts of interest).
I think it's because they've already accepted the fact that all of their trading data is moving through those terminals and have that reality figured into their mental model of how their business operates. I don't think it's figured into their risk models however.
We also utilize Bloomberg Anywhere that let's you login remotely from anywhere with a fingerprint scanner.
They don't have total vertical integration. Rarely do they own the data - most of it comes from other parties, either contributions from broker-dealers or from exchanges. Thats why you have to buy the special data sets on bloomberg. They do a lot of work digitizing quotes but these days its much less important than it used to be.
There are many companies (most significantly Markit) who are looking to crush this beast. Reuters has already failed IMO. But, the important thing is with the fragmentation of trading venues there's been fragmentation of the market data business. Bloomberg is great at aggregating many different data sources, but there are others (e.g. the NYSE) that could provide a great service that is geared towards a specific client niche. I think that's the more likely danger to them.
Times have changed, though, and the expectation now is that users have absolute privacy from their service provider except on a need-to-know basis, and Bloomberg is going to have their work cut out for them adapting their systems and culture.
The problem is that Bloomberg sells data services to tens of thousands of companies while also running a media company that sells reporting to news organizations. You can't mix those two and expect people to be happy about it.
Whereas reporters from other news organizations would have had to ring around and fight to speak to a relevant source, Bloomberg reporters had direct contact details at their fingertips, as well as information telling them exactly when they are logged in at their Bloomberg terminals.
The customers might very well have had an expectation of privacy based on the contracts they signed with Bloomberg. Given that Bloomberg's customers are financial companies that have lots of proprietary information to protect, I'd be surprised if their contracts didn't contain some kind of confidentiality agreement.
No credentials were exposed.
What's the issue?
It had been speculated that Bloomberg employees were looking at the actual content of searches, which would be a death blow to their reputation as a trusted platform. This story is Bloomberg saying "Yes, we gave access to info to employees we shouldn't have, but it was actually pretty banal."