[Desk Column] Two Reasons for Opposing the Data Agency

Does watching violent movies really make people more violent? Two American economists, Gordon Dahl and Stefano DellaVigna, analyzed big data from 1995 to 2004 to find out. They examined hourly crime data from the Federal Bureau of Investigation (FBI), movie box office rankings, and movie violence ratings. As a result, they discovered that crime rates actually dropped on weekend evenings when violent movies were shown. Big data said this: "If people with latent violent tendencies watch movies instead of gathering to drink, violent incidents can decrease."

Seth Stephens-Davidowitz, a data analyst at Google, listed other big data success stories in his book Everybody Lies. Walmart uses nationwide store sales data to decide which products to place on shelves; Netflix recommends movies based on the clicks and views of doppelganger customers; and although polls predicted defeat, Google searches foresaw Trump's victory in the 2016 U.S. presidential election...

As the book title suggests, people may lie when responding to polls, but they do not lie when searching on Google alone in their rooms. This is the "value of data." The scholar Yuval Harari also praised the value of big data with the term "Dataism" in Homo Deus. "This universe is made up of flows of data, and its supreme value is the 'flow of information.'"

Recently, there has been a competitive surge of calls in the political sphere to establish a "Data Agency." The reason is the need for a control tower for data policy. This necessity is fully understandable. Data is the "oil of the future," and as Yuval Harari said, we must also participate in the "cosmic data flow." Yet, why does the idea of a Data Agency feel unappealing?

First, because the Data Agency proposed by politicians is more likely to rely on regulation rather than promotion, and control rather than autonomy. The creation of a new government organization means budget and personnel are allocated, and authority to manage and supervise laws is granted. Of course, appropriate regulation is necessary when the market is chaotic. But now is the time to take out the dusty data accumulated under regulatory frameworks and assign economic value to it. This is why companies were indifferent when the Data Agency idea was raised. There is no need to mention former U.S. President Ronald Reagan's remark that "government is not the solution to our problem; government is the problem." Regulations are easy to make but hard to abolish.

Another reason to oppose the Data Agency is that this is a matter of strategy, not execution. An agency is an executing body, not a place to formulate strategy. Data-related policies are already scattered across the Ministry of Science and ICT, the Ministry of the Interior and Safety, the Financial Services Commission, and soon the Personal Information Protection Commission, which is responsible for personal data protection, will be launched. At a time when macro and comprehensive consideration is needed on how to coordinate these functions, execution without strategy is worse than doing nothing.

Then what about other countries? In so-called big data advanced countries, data policy is not concentrated in a single ministry. Data policy is a national governance task of "coordination and cooperation" that cannot be monopolized by a specific department, as seen in the UK, where each ministry implements policies reflecting the characteristics of their data, or in the U.S., where ministries collaborate through a big data council.

Some argue that a committee under the president or prime minister should be created to coordinate whole-of-government policies, but the same applies. It must not become redundant regulation on top of regulation, and a macro and comprehensive strategy must be established. Is this really possible? Yuval Harari provides the answer: "Capitalism won over communism because the flow of information was freer. Thanks to the freedom of information. Therefore, the key to creating a better world is to freely release data."