Pubblicazioni

​

​

At every level of government, officials contract for technical systems that employ machine learning—systems that perform tasks without using explicit instructions, relying on patterns and inference instead. These systems frequently displace discretion previously exercised by policymakers or individual front-end government employees with an opaque logic that bears no resemblance to the reasoning processes of agency personnel. However, because agencies acquire these systems through government procurement processes, they and the public have little input into—or even knowledge about—their design or how well that design aligns with public goals and values. This Article explains the ways that the decisions about goals, values, risk, and certainty, along with the elimination of case-by-case discretion, inherent in machine-learning system design create policies—not just once when they are designed, but over time as they adapt and change. When the adoption of these systems is governed by procurement, the policies they embed receive little or no agency or outside expertise beyond that provided by the vendor. Design decisions are left to private third-party developers. There is no public participation, no reasoned deliberation, and no factual record, which abdicates Government responsibility for policymaking. This Article then argues for a move from a procurement mindset to policymaking mindset. When policy decisions are made through system design, processes suitable for substantive administrative determinations should be used: processes that foster deliberation reflecting both technocratic demands for reason and rationality informed by expertise, and democratic demands for public participation and political accountability. Specifically, the Article proposes administrative law as the framework to guide the adoption of machine learning governance, describing specific ways that the policy choices embedded in machine-learning system design fail the prohibition against arbitrary and capricious agency actions absent a reasoned decision-making process that both enlists the expertise necessary for reasoned deliberation about, and justification for, such choices, and makes visible the political choices being made. Finally, this Article sketches models for machine-learning adoption processes that satisfy the prohibition against arbitrary and capricious agency actions. It explores processes by which agencies might garner technical expertise and overcome problems of system opacity, satisfying administrative law’s technocratic demand for reasoned expert deliberation. It further proposes both institutional and engineering design solutions to the challenge of policymaking opacity, offering process paradigms to ensure the “political visibility” required for public input and political oversight. In doing so, it also proposes the importance of using “contestable design”—design that exposes value-laden features and parameters and provides for iterative human involvement in system evolution and deployment. Together, these institutional and design approaches further both administrative law’s technocratic and democratic mandates.

Algorithms are among the most important technological drivers of the ongoing digitalization process. They are becoming more and more important, enabling firms to be more innovative and efficient. However, debate has arisen on whether and to what extent algorithms might also have detrimental effects on the competitive functioning of markets. In their joint conceptual project – Algorithms and Competition – the Autorité de la concurrence and the Bundeskartellamt studied potential competitive risks that might be associated with algorithms. They elaborated on the concept of algorithm as well as on different types and fields of application. In their study, the two authorities focused in particular on pricing algorithms and collusion, but also considered potential interdependencies between algorithms and the market power of the companies using them as well as practical challenges when investigating algorithms. Isabelle de Silva, President of the Autorité de la concurrence: “Algorithms are used constantly in the digital economy, and are at the very core of how some fast growing businesses operate: online travel agencies, e-commerce, online advertising, to name only a few. It is essential that we look into how these algorithms work. We need to determine if there is a risk that algorithms might facilitate or permit behaviours that are contrary to competition law. With this joint study with the Bundeskartellamt we aim at reaching a common view on these matters and at starting a debate with stakeholders.” Andreas Mundt, President of the Bundeskartellamt: “The joint study is another proof of the continuing cooperation between our agencies. As digital markets keep evolving, we expand our expertise on algorithms in an exchange with each other. This is in line with our efforts to devote more resources to the digital economy with the clear-cut aim to enforce competition law also in the era of platform economy and digital business models.” Algorithms and competition are also the topic of an accompanying conference hosted by the Autorité de la concurrence and the Bundeskartellamt that is taking place in Paris today. Several renowned speakers, including business representatives, researchers and competition enforcers, are discussing potential business applications for algorithms, pricing algorithms and the risk of horizontal collusion, as well as ways to address the challenges raised by algorithms.

​

The world’s largest businesses must routinely police other businesses. By public mandate, Facebook monitors app developers’ privacy safeguards, Citibank audits call centers for deceptive sales practices, and Exxon reviews offshore oil platforms’ environmental standards. Scholars have devoted significant attention to how policy makers deploy other private sector enforcers, such as certification bodies, accountants, lawyers, and other periphery “gatekeepers.” However, the literature has yet to explore the emerging regulatory conscription of large firms at the center of the economy. This Article examines the rise of the enforcer-firm through case studies of the industries that are home to the most valuable companies, in technology, banking, oil, and pharmaceuticals. Over the past two decades, administrative agencies have used legal rules, guidance documents, and court orders to mandate that private firms in these and other industries perform the duties of a public regulator. More specifically, firms must write rules in their contracts that reserve the right to inspect third parties. When they find violations, they must pressure or punish the wrongdoer. This form of governance has important intellectual and policy implications. It imposes more of a public duty on the firm, alters corporate governance, and may even reshape business organizations. It also gives resource-strapped regulators promising tools. If designed poorly, however, the enforcer-firm will create an expansive area of unaccountable authority. Any comprehensive account of the firm or regulation must give a prominent role to the administrative state’s newest gatekeepers.

The draft OECD Best Practice Principles for Regulatory Policy: Reviewing the Stock of Regulation [the Principles] is hereby presented for public consultation. The aim of the Principles is to provide policymakers and civil servants in both OECD member and partner countries with a practical instrument to better design and operationalise their ex post evaluation systems. The objective is to complement the 2012 Recommendation on Regulatory Policy and Governance which, albeit clear on the importance of undertaking evaluations in the overall context of the regulatory lifecycle, is rather general on providing specific guidance. In these Principles, particular attention is paid to ensuring that regulation is comprehensively reviewed over time, and that the reviews include an assessment of the actual outcomes achieved against their rationales and objectives. The Principles further expand on these areas by covering both the overall governance of ex post evaluation systems and individual reviews, various methodologies that can be used as assessment tools, public consultation throughout the review process, as well as capacity building and leadership.

​

Over the last decade the Open Government Data movement has successfully highlighted the value of data and encouraged governments to open up information for reuse both inside, and outside the public sector. This Working Paper argues that governments now need to go further and put the role and value of data at the core of thinking about the digital transformation of government. A data-driven public sector (DDPS) recognises that data are an asset, integral to policy making, service delivery, organisational management and innovation. The strategic approach governments take to building a DDPS can have a positive impact on the results they deliver by promoting evidence-led policy making and data-backed service design as well as embedding good governance values of integrity, openness and fairness in the policy cycle.

​