Pickles Unpicks CESR's Level 3 Guidance on MiFID's Data Challenges

The Committee of European Securities Regulators recently issued its Level 3 guidance and recommendations concerning market data challenges resulting from the European Commission's MiFID regulations. Chris Pickles, industry relations manager at BT Radianz and chair of industry body the MiFID Joint Working Group, spoke to Inside Market Data to explain the reasons behind each recommendation.

IMD: CESR's first guideline requires that data be "monitored continuously for errors," using a system that provides an "independent cross-check." Why was this necessary?

Pickles:

Market regulators in some countries wanted the data published by investment firms and new trading platforms to be monitored in the same way that stock exchanges have traditionally monitored data. The argument being put forward by some exchanges was that if the regulators allowed anybody to publish this data, there would be a loss of quality.

In the UK, the FSA was saying it wanted to have approved aggregators of data who actually quality check. But there has been some push-back from data vendors, who have said... "We are commercial entities, we're not regulated and we already carry out quality checks on data because… if we sell bad data it reflects badly on our business." They don't want to take on that role as a regulated body. Equally, investment firms themselves have said, "We want to publish data directly to the market, we don't want to go through a third party,"—and MiFID allows that.

So guideline one brings in the principle that… [quality controls] have to start with the source of the data, and it has to be an independent cross-check, but you don't need to go to a third party to do that.

IMD:

The second guideline concerns duplication of post-trade information. What were the issues here?

Pickles:

MiFID allows an investment firm to publish directly to the market or through third parties. It's logical and reasonable that an investment firm might sell its data via 20 different data vendors, and each data vendor may identify an individual trade report with a different identification system. So… how do you know which transactions are duplicates of each other?

Where people need to eliminate double counting—and that's not everybody—then you need to know where you can go to an original unique source. What's likely to happen, one hopes, is that the identification number used by whoever is the primary source, will continue to be used [throughout the chain]. With guideline two, the idea is that you should at least nominate a primary source so that if you do have an issue around duplicate data, you can [choose to receive data only] from vendors who either get data directly from a primary source or who are themselves a primary source. As a contributor of data, you can have different primary sources for pre- and post-trade data and for different instruments.

An original suggestion to use time stamps wasn't accepted… largely because… time stamps would be one way of identifying something, but there is a principle that you don't use something that has one meaning to identify something else. Time stamps mean time, so they shouldn't be used as an identifier.

IMD:

Guideline three suggests that the ultimate responsibility for compliance "resides with the relevant regulated market, MTF or investment firm." Does this just mean you can't pass on legal liability for your transparency requirements?

Pickles:

Right. It's like those posters that say, "This means you!" You might outsource it, but ultimately it's your responsibility.

IMD:

Guideline four concerns contingency arrangements for publishing data, which most exchanges surely already have in place. What's the point of this?

Pickles:

MiFID actually has business continuity requirements in article 13 saying that you need to have backup facilities and you have the responsibility to keep running even if there are troubles with your technology.

IMD:

The use of Web sites to publish data is interesting. Will that actually be a popular way of distributing data?

Pickles:

If you're looking for the lowest-cost way of publishing data, then dumping something on your Web site is perceived by most as the easy way of making data available to the market as a whole. But the difficulty is then how do you put it on a Web site [in a way that will] make it more or less accessible and easy to consolidate? We tried to propose words that were as close as you can get without specifying the exact format. The idea of it being machine-readable means that it wasn't designed to be read by a human being. One of the difficulties in explaining this to regulators was differentiating the Internet from Web sites. A broker can pump a stream of data over the internet as a datafeed, but the Internet isn't the same as a Web site. If you start using the word Web site, you have to define what you mean. If you print a deal ticket as a PDF and put it on a Web site, it can only be read by a human. This all goes back to the principle that the data should be easy to consolidate.

We wanted it to [refer to] feeds to propose the idea that it should be pushed out, but the regulators didn't go as far as accepting that. So [what was accepted was] the idea that you could put it on a Web site, but it had to be somewhere you can find it, and in a format known in advance [that is] structured and machine-readable. Ideally, you would do it as a datafeed. If you want to make money from it, you should push it out.

IMD:

The sixth guideline relates to timing of post-trade publication. Is CESR saying that there's no excuse for not publishing in real time?

Pickles:

No longer than three minutes is the rule. But CESR is saying that if you do publish close to the three-minute line regularly, then they'll be looking at you. If you weren't able to do it in real time, or very close to execution, then the technology would be counted as inaccurate, and you'd have to change that.

IMD:

Guideline seven talks about the availability of information and says that data must be made available on a non-discriminatory and reasonable commercial basis. What does non-discriminatory mean, and how do you benchmark reasonable commercial terms?

Pickles:

The element of non-discriminatory is to make sure you're not cutting different deals unfairly. You might cut one deal with a data vendor that will sell it to a 100,000 customers, and a different deal with Chris Pickles, who just wants it for his own use. Also you've got to think about geographies. It says that any investor in "the community" should be able to access data on a non-discriminatory basis. I'm not aware that means the European Community; I think it means the [global] financial community.

One of the benchmarks in measuring this is likely to be the price of exchange data. With market forces, if your data is unimportant, don't expect people to buy it from you. The opposite extreme is if you are Project Boat and you have the best data in the market, in which case you have a much better negotiating position and can charge what it's worth.

IMD:

What is the purpose of not allowing bundling of information services?

Pickles:

Let's say you are an exchange and you decide to be a data aggregator. You collect MiFID data from firms to publish on their behalf, and you charge a reasonable commercial price for that MiFID data. If somebody comes to you and says "I only want the MiFID data," you'd have to do that. Also, that doesn't just apply to data sets. It could also apply to bundled services, like network provision.

IMD:

Finally, CESR recommends the use of standards in publishing data.

Pickles:

With standards, CESR has recognised industry participants' wish that standards are used where possible. For example, they recommend the use of ISO standard formats for date and time, and the use of ISINs. The use of Bank Identifier Codes (BICs) to identify trading counterparties is somewhat more of a problem because not all trading counterparties have a BIC.

Only users who have a paid subscription or are part of a corporate subscription are able to print or copy content.

To access these options, along with all other subscription benefits, please contact info@waterstechnology.com or view our subscription options here: http://subscriptions.waterstechnology.com/subscribe

You are currently unable to copy this content. Please contact info@waterstechnology.com to find out more.

Enough with the ‘Bloomberg Killers’ already

Waters Wrap: Anthony interviews LSEG’s Dean Berry about the Workspace platform, and provides his own thoughts on how that platform and the Terminal have been portrayed over the last few months.

Banks seemingly build more than buy, but why?

Waters Wrap: A new report states that banks are increasingly enticed by the idea of building systems in-house, versus being locked into a long-term vendor contract. Anthony explores the reason for this shift.

Most read articles loading...

You need to sign in to use this feature. If you don’t have a WatersTechnology account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an individual account here