Approximately 51% of the adult population in Indonesia remains unbanked. They cannot fulfill formal requirements to prove their credit worthiness and are therefore not considered for services by conventional financial institutions. Addressing this barrier to their financial well- being, innovative credit scoring (ICS) has emerged as a solution for financial inclusion. Digital service providers have developed tools to gather alternative relevant data that enable them to service a much wider segment of society.

However, the promise of ICS to extend financial inclusivity to the underserved population is not without caveats. Key risks include data inaccuracy, a lack of data privacy, heightened exposure to cyber risks, and potential for increasing or entrenching discrimination. As with any emerging digital financial innovation, further clarity in regulation, technology use, and data protection is needed.

While Indonesia’s ICS sector is still new and developing, it has been widely used for years in some developed markets including the United States and China. Following a CIPS policy paper on “The Rise of Innovative Credit Scoring Systems in Indonesia,” this discussion paper unpacks the approaches to credit scoring in China and the United States to draw lessons for Indonesia.

Looking at the Chinese experience, it is fair to say that excessive use of data and transparency issues remain key constraints upon a robust rating system. The Social Credit System that has been put in place alongside ICS, with widespread data-sharing to third parties, indicates the need for a clear-cut accountability chain. The state is in a difficult, maybe an impossible position, in which its interest in gathering data and concealing how it is used is in conflict with its obligation to its citizens to regulate and control the associated risks to privacy, cyber security, and users’ ability to control their own data.

Meanwhile, the experience with the ICS industry in the United States can inform the development of effective policies and practices in Indonesia. In the United States, the relevant policy discourse focuses not just on market power and data privacy, but also on algorithmic governance and socioeconomic biases. In particular, concerns about socioeconomic biases and algorithmic decision making has not yet gained traction in China and Indonesia, which both have substantial minorities that might find themselves subject to discriminatory access to financial services through AI and machine learning biases.

Observing these international experiences helps us better understand the risks and challenges associated with innovative digital solutions that appear and evolve at a rapid pace in Indonesia.