Nonrivalry and the Economics of Data

Von Ralf Keuper

Bei der Frage, wie der ökonomische Wert der (personenbezogenen) Daten zu bemessen und wie sinnvoll ein Eigentumsrecht an Daten ist, gehen die Meinungen auseinander. Eine etwas andere als die sonst übliche Betrachtungsweise wählen Charles Jones und Christopher Tonetti in Nonrivalry and the Economics of Data. Dabei orientieren sie sich u.a. dem Konzept der Idea Economy von Paul Romer (Vgl. dazu: Er ist der Prophet des ungebremsten Wachstums. Immer noch. Der Ökonom Paul Romer über Produktivitätssteigerungen dank nie versiegender Ideen und die sinnvolle Subvention von Bildung). Im vergangenen Jahr erhielt Romer den Nobelpreis für Wirtschaftswissenschaften.

Romer (1990) emphasized that ideas are nonrival. We add data to this taxonomy, .. . We find it helpful to define information as the set of all eco- nomic goods that are nonrival. That is, information consists of economic goods that can be entirely represented as bit strings, i.e., as sequences of ones and zeros. Ideas and data are types of information. An idea is a piece of information that is a set of instructions for making an economic good, which may include other ideas. Data de- notes the remaining forms of information. It includes things like driving data, medical records, location data, and consumption history that are not themselves instructions for making a good but that may still be useful in the production process. An idea is a production function whereas data is a factor of production.

Über die Unterschiede zwischen Ideen und Daten:

The data and the idea are distinct: the software algorithm is the idea that is embedded in the self-driving cars of the future; data is an input used to produce this idea. Once the model is estimated, the data can be thrown away. …. Engineers change jobs and bring knowledge with them; people move and communicate causing ideas to diffuse, at least eventually. Data, in contrast, especially when it is “big,” may be more easily monitored and made to be highly excludable. The “idea” of machine learning is public, whereas the driving data that is fed into the machine learning algorithm is kept private; each firm is gathering its own data. … A new idea is a new production function for producing a variety while data is a factor of production.

Die eigentliche Fragestellung:

When firms restrict the sale of data to limit their exposure to creative destruction, what are the consequences? When consumers own data and can sell it, is the allocation optimal? What if data sharing is banned out of a concern for privacy?

Firmen, die den Zugang zu ihren Daten beschränken oder verhindern, schaden letztlich nur sich selbst, sonder auch der gesamten Wirtschaft eines Landes:

In an economy in which firms do not share data, firms learn only from their own production. … Contrast this with an economy in which data is shared. In that case, the amount of data that each firm can learn from is an increasing function of the size of the economy. Therefore, the scale of the economy and the increasing returns associated with the nonrivalry of data interact.

Entscheidend für den Erfolg einer Volkswirtschaft in der Datenökonomie ist nicht die Menge der Daten oder die Größe eines Landes, sondern die Fähigkeit, den Anschluss an den weltweiten Ideen- und Datenstrom sicherzustellen:

For example, while ideas give rise to increasing returns and people create ideas, growth theory does not typically suggest that Luxembourg and Hong Kong should be much poorer than Germany and China because of their relatively small size. Instead, the view is that ideas diffuse across countries, at least eventually and in general, so that the relevant scale is the scale of the global market of connected countries rather than that of any individual economy.

Aus Sicht der Allokation, der effizienten Ressourcenverteilung, sehen die Autoren in einem Eigentumsrecht der Nutzer an ihren Daten kein Hindernis für die Datenökonomie und auch keine negativen Konsequenzen für die Personen wie überhaupt die Gesellschaft:

Instead, our analysis indicates that giving data property rights to consumers can lead to allocations that are close to optimal. Consumers appropriately balance their concerns for privacy against the economic gains that come from selling data to all interested parties.


Dieser Beitrag wurde unter Datensouveränität, Identity Economy veröffentlicht. Setze ein Lesezeichen auf den Permalink.