Co-regulating algorithmic disclosure for digital platforms

Abstract

With digital platforms gaining dominant intermediating role and exerting regulatory functions vis-à-vis small and medium-sized enterprises (SMEs) through algorithms, EU institutions have started considering to rely on their analytical capacity to regulate the myriads of market transactions occurring within and through them (so-called platform-to-business, or P2B transactions). Most of the time, the EU suggests recurring to light-tough disclosure duties. Hence, the European model falls short in rebalancing information asymmetry and unequal bargaining power plaguing the SMEs. In practice, the EU model consists either in pure delegation of self-regulatory powers (codes of conduct) or non-enforceable co-regulatory schemes (with technical standards established by the platforms themselves). Other models have been suggested that rely on the regulator's access to the platform's data (so called savvy and data-delegated options). These governance models present several limitations, making the platforms' role as regulatory intermediators little credible. In this scenario, the paper purports that a third option should be considered. In particular, to tackle the multifaceted risks associated with algorithmic decisions by digital platforms, while at the same time avoiding stifling innovation, it makes three suggestions: (1) also information disclosures should be done by an algorithm; (2) that is pre-tested in a co-regulatory process, that involves the regulator and stakeholders; and (3) enforced through legal and other empowerment tools, rather than sole fines.

Keywords

• digital platforms
• big data
• p2b
• smes
• self-regulation
• co-regulation
• disclosure
• algorithmic co-regulation
• terms and conditions
• regulatory intermediation
• sandbox
• knowledge graph