Data Ownership and the Community Platform: Use Case
hen we started work on our fellowship, remote work was a niche concept. It has now become the norm in large organisations such as Stripe, Facebook and Airbnb. We used to think last-mile delivery personnel in the gig economy are not essential to the functioning of a society. However, with the pandemic and lock-downs, they became crucial for delivery of multiple goods. The year of the pandemic saw accelerated changes from a policy and technical standpoint, though more work is needed to accommodate these paradigm shifts. In the final piece of our fellowship, we present some of the critical concerns on the state of the gig economy today, as precursor for the data and final paper that we will publish by early May 2021.
According to the Fairwork India Ratings report of 2020, workers on seven of the eleven prominent gig economy platforms had insufficient evidence of paying minimum wage standards for employee income on their platform. This is paradoxical because most gig workers join platforms such as Swiggy or Uber expecting higher earnings. The qualitative data from our surveys showed that a high percentage of these workers are usually locked-in due to investments made to move to a city or purchase physical assets (e.g. bikes, cars) that facilitate their transition to the gig economy. Today, most individuals when they enter the gig economy see it as a source of lifetime income.
As seen in the recent imbroglio of a Zomato employee’s alleged misbehaviour with a social media influencer, building up their reputation can stand gig economy workers in good stead. The record of past performance, delivery times and favourable reviews from customers mean greater credibility for the gig workers when something adverse crops up. The challenge here is that the platform holds the right to be neutral and not disclose these figures. A publicly verifiable reputation system in this context would have removed the individual’s dependency on the platform to reveal these statistics. Likewise, if he were a member of a community/co-operative/digital union, his credibility would have been higher, as the group that he was part of would have vouched for him. By summarising our research until now, we shall now look at how such a system can be built.
In our previous blog, we had alluded to a community platform as a possible solution, keeping the worker’s identity as its core. Workers are incentivised to become members of multiple communities (Skill based, Health based etc) as such membership enhances their wellbeing. Given below is a representation of how these incentives are positioned as determinants of the workers’ health, mental and financial wellbeing.
Ideally, with this portable skill identity, which is vetted and validated by the communities they are part of, the gig worker is able to upskill and move his or her skill across jobs / gigs that leads to a reputation score. This score which includes validated skill development and skill movement data from the co-operatives/ communities would lead to group financial services (sachet insurances, mutual funds, banking etc)
However, to ensure that any community platform does not turn into another big-tech platform, we suggest a model of self-sustaining communities with data ownership at the gig worker level and skill validation and skill movement transaction data at the community level.
The key aspects of the data cooperative/community:
Individual members own and control their personal data 
Fiduciary obligations to members: First and foremost, the data cooperative has a legal fiduciary obligation to its members . The organisation is member-owned and member-run, and it must be governed by rules (bylaws) agreed to by all the members.
Direct benefit to members: The primary goal of the data cooperative is to benefit its members. The goal is not to ‘monetise’ their data, but to perform on-going analytics to understand the needs of the members better and to share insights among the members. 
An overview of the data cooperative (Community) ecosystem is below. Backed by the 4000+ gig worker interviews we have conducted in the last 4 months, in our final thesis, we would be exploring the system and policy architecture for an open source, skill-identity-based reputation score that is user-owned and governed through a consortium. This consortium includes public, private and social institutions and is distributed, decentralised and validated through communities that the gig worker is part of to upskill and move their skills.
Our data shows that gig platform workers wish to upskill themselves, but are constrained by lack of financial resources and time, since they need to put in 12–14 hours every day. Although fintech applications now offer insurance, loans and savings deposits for gig workers, they are not educated on the intricacies of financial literacy. Likewise, the mental health challenges that arise from being engaged in the gig economy are hardly addressed. Many of the respondents take help of Instagram or Tiktok to find relief from the monotony and tedium of their work. While technology-based counselling can help, a shift in the gig workers’ conception about their mental health can occur only if there is meaningful involvement by the platforms that employ them. Since the gig economy is set to employ over 70 million individuals within the next decade, this assumes great importance. As a percentage of the labour force, it could soon compete against other sectors such as servicing and manufacturing. The gig economy faces many challenges today, ranging from upskilling to a brewing mental health crisis. We are witnessing the failure of the state and markets to cater to the needs of the average gig worker. The United States and other developed countries have started to allow contract workers to form unions, e.g. in Amazon. It is only when new resources such as personal data are leveraged that the average worker truly finds empowerment. As we have suggested in the past, a holistic gig worker-oriented solution should have personal data as its foundation. On top of this, interoperability of professional reputation and transaction data could be merged to create a single API that allows potential employers to find workers and vice versa.
Y. A. de Montjoye, E. Shmueli, S. Wang, and A. Pentland, “openPDS: Protecting the Privacy of Metadata through SafeAnswers,” PLoS ONE 9(7), pp. 13–18, July 2014, https://doi.org/10.1371/journal.pone.0098790
Roles of a data cooperative: J. M. Balkin, “Information Fiduciaries and the First Amendment,” UC Davis Law Review, vol. 49, no. 4, pp. 1183–1234, April 2016.
Data Cooperatives by Alex Pentland and Thomas Hardjono Apr 2020 -