There’s been lots of interest in Dr Carla Bonina’s recent interview on Open Data in a piece published at the Guardian in December. Below, Carla expands on further points from her field-leading research.
Open data continues to be at center of stage in Europe and the countries that support the Open Government Partnership (OGP). Last year, the Banque of France—the country’s Central Bank—open up a lab that would give access to millions of records to accredited researchers to conduct research. That was just one of the examples in France’s agenda on government openness before hosting the annual OGP summit in Paris.
What’s open data? In a nutshell, we call open data any piece of digital, machine readable data that anyone is free to use, reuse, and redistribute. Open data can be small, large or medium – so there’s no contradiction, in principle, between the concept of open data, and big data. If you think of downloading the whole of Wikipedia, that’s an example of open data that’s also big data. In the banking sector, bear in mind the volume of transactions; it follows that the volume of data is generally very big. But as there’s a large amount of personal data involved, the amount that’s open in this area is very limited.
I believe that a significant element of the banking industry has lobbied for this level of secrecy. But other sectors that use sensitive data – such as health – have been more ready to open up than financial services. One of the hottest controversies in the world of research is that if data is closed, you don’t have many possibilities to replicate studies — and if you can’t assess the outcomes of research, you are compromising one of the key principles of science. In health, after a controversy with drug testing, the British Medical Journal launched an open data campaign, in which they will no longer publish trials of drugs where authors do not commit to making relevant patient level data available in open format. So, we see how the move to open data makes a difference. If you look at any financial or economic model that might emerge from banking data, you could argue that someone in future should have access to the same data. It’s more available for being checked and verified than many models we have at the moment.
In the commercial sector, we see the move towards more open data similarly contested. In most countries, central banks are one of the biggest producers of data – which is usually high quality, closed statistics rather than open data. Even when that data is anonymized, experiments show that you can always de-anonymise data. It’s a sensitive issue: regulators in the financial sector may be inclined towards open data (e.g. the Bank of England), but the risks involved are high. For example, people could be targeted for cybercrime, or charged higher insurance premiums according to income levels (despite the latter not being necessarily legal). You have to be very sure that you’re not going to make a policy mess through the disclosure of personal data that can harm the rights of individuals.
The recent move by the Banque de France towards opening up data — although with its own restrictions — is a healthy sign that this traditionally secretive sector is not going to be excluded from the open data movement. Privacy and confidentiality are important concerns – and who benefits from the data is also a major issue. That won’t go away. However, proper open access to data has potential not just in terms of the economic benefits that could be created – for example, through more competition, and better products and services for consumers in the financial services. But more importantly, open data in the financial services can help guard against corruption in financial services—and if you need reminding why this is important, it’s worth checking out the money laundering and bribery scandals that the Panama Papers put forward last year.
Stay tuned to CoDE to read more views from our experts on the hottest topics stirring up the Digital Economy!