Market Data Management: Financial information as a business resource

Vendors in the Asia-Pacific region have more than doubled their spending on market data between 2000 and 2008.
In 2008 the estimated total Asia-Pacific market data spend was $3.423 billion – this enormous figure leads to the question: Is market data being used efficiently in Asia?

Key points of the session:

• Data efficiency is still not achieved.
• Lack of consistency in both legal and technical areas is a major issue.
• Data validation is a time-consuming problem.
• Lack of universally adopted data identifiers is holding the market back.
• Low-latency in conjunction with accuracy of data is the providers’ mission.

Speakers:

• Keiren Harris [Mod] – Panel moderator – Market Data consultant
• Robert Jeanbart [RJ] – Founder of SunGard MarketMap and Global Head of Market Data and Information Services, SunGard Global Trading
• Will Mateu [WM] – Business Development and Sales, ICAP Market Data Services, Australia and Asia
• John Pilch [JP] – Sales Director, Asia-Pacific, TNS
• Dan Ryan [DR] – Director and Legal Counsel for innovative desktop and mobile and market data provider, Carryquote
• Michael Wong [MW] – Associate professor, City University, providing quantitative risk analysis for banks in Hong Kong

Is market data being used efficiently in Asia?

RJ – Market data is a big ‘pain point’ for organisations anywhere because data ranks no.2, if not no.1 on the total spend by financial institutions. There are so many vendors, sources, and types of data that it is hard to use them all efficiently – it is painful, costly, and labour-intensive.

MW – In large banks data comes from a variety of sources, but they have data management centres to handle it. Smaller institutions may rely on fewer sources of information and can have a problem because incorrect data will affect pricing, variation and risk control – a mechanism is required to check and verify the data.

DR – Efficiency of data means that end-users are able to extract the value they need, from the prices they can afford, in the way that they want to consume data. That is what our company is trying to do – allow users that are interested in real-time data to do so at the price-point they can afford. I don’t think that’s necessarily well covered in Asia or other markets at the moment.

JP – Clients haven’t been getting real-time data; they’ve been getting the normalized feeds which are obviously delayed, there’s no efficiency there. However, lately we have seen customers asking for data straight from the exchange. Also, vendors themselves are coming to us and wanting best latency – everything is heading towards efficiency but there’s still a way to go.

WM – At ICAP we’re both a consumer of data and a provider out into the wider market. We’re look at, not only the efficiency, but the continuity of getting information through to front-, middle- and back-offices. As an independent source of information, we want to facilitate the clients’ use of our content in an efficient way.

Consistency is certainly one of the issues.

DR – I wanted to focus on the lack of consistency in the licensing terms amongst the different sources of data. Why should there be such differences in licensing models between exchanges providing more or less the same data. Why should the Hong Kong and Singapore exchanges offer real-time snapshot data but the Taiwan and Tokyo exchanges do not?

RJ – Yes, now the ‘pain point’ turns into a nightmare…We are acquirers of 160, maybe 170 data sources and I can almost guarantee that no two of them have similar pricing or reporting mechanisms.

JP – We also have a lot of problems actually connecting to the various exchanges because of the different legal and also technical issues. In terms of the question, yes, we have a lot of problems with consistency. In fact, nothing is consistent at this stage.

Is data validation a problem?

RJ – I can give you a statistic: for every hour we spend acquiring a new data source we spend four more hours cleansing that data.

DR – We have that problem too. I couldn’t quote the precise ratio but that sounds correct; but verification is an obligation – you have to ensure that data from your system is correct.

And data identifiers are also an issue?

RJ – Symbology, recently promoted by Bloomberg, could change the face of the industry; if universally adopted, this would be very good news.

MW – Currently, identifiers can have different meanings in different sources: this might not immediately be noticed.

WM – We just adapt to the different standards. As brokers we are just facilitators, we can’t impose our coding on data suppliers.

RJ – I don’t think it is that simple. There is the instrument identifier and then there are the field identifiers – these can number as many as 1500. It is a big issue which needs to be standardized.

How real an issue is low-latency?

JP – It is very real. Everyone wants the lowest latency. However, not all exchanges cater for low latency – some still batch trade and price data.

MW – Yes, data timing is crucial.

RJ – Price discovery is where you need to be incredibly fast but the fundamental equation is still that between speed – latency – and the correctness of data. If an algorithm is fed with bad data the results can be disastrous. We have invested massively for low latency but this has always been in conjunction with accurate data.

What would you like to see your suppliers of data doing better?

MW – Data integrity is very important in building trading and risk control models. Providers should employ monitoring systems to guarantee the integrity of data or else all your decisions, all your support systems, no matter how expensive, will be useless.

RJ – Clients that have structured their business and workflow generally have efficient data management. But business, and the variety of instruments, does grow. Effective, ongoing planning of data management is important.

Lastly, if you had to give one piece of advice to an institution as to what it should be doing regarding future data requirements, what would it be?

JP – Reduce the complexity of data overheads. Reduce your number of carriers and vendors to achieve your aims at the right price, in a timely manner, with stability.

WM – Heatmap data-usage across the business to determine what is and is not being utilized.

MW – Invest in good data warehousing to ensure the capture and verification of data.

RJ – If I were in my clients’ shoes I would go through a process to determine what is crucial to my business; what is my domain, my expertise. Any thing that falls outside I would out-source. I would solve that equation of the right data, at the right price, at the right usage.