(P1) Algorithmic Regulation: What are Regulators Doing with Data and Can They Make Better Use of It?
David Demeritt, Professor of Geography, King’s College London
Henry Rothstein, Reader in Risk and Regulation, King’s College London
How can regulators capitalise on the ‘Big Data revolution’? If Amazon and Google can use big data analytics to predict our personal habits and beliefs, can regulators do something similar to predict regulatee behaviour and non-compliance? Certainly, as the volumes of data generated as a digital by-product of everyday transactions and administrative routines continue to grow, so does interest in new forms of ‘algorithmic regulation’ as a means of delivering faster, efficient and more effective regulatory action (O’Reilly 2013; Yeung 2018). Yet while regulators increasingly claim to want to ‘make better use of data’, little is known about the use of such approaches by regulators, how and why they are used, nor, indeed, their effectiveness.
This panel will start to fill that gap with a series of papers mapping the use of ‘algorithmic regulation’ in healthcare and higher education quality regulation, where many of the leading-edge developments are taking place. Drawing on extensive international research, the papers will examine the drivers behind algorithmic regulation in these domains, consider how and why its design and use varies from country to country and identify significant challenges. The panel will also include a presentation of the first ever peer-reviewed test of whether existing algorithmic regulatory techniques are statistically reliable or even possible in principle in these domains, demonstrating that past approaches claimed to be world-leading were in fact failing, but also pointing to how the application of machine learning tools to social media may be more successful (Beaussier et al 2016; Griffiths et al 2017, Griffiths and Leaver 2017).