Giant Economy Monitor


How economies work is poorly understood by most citizens – how corporate and political interlocks shape industrial policies or resource allocation, how political alignment of elected representatives affects the growth of their constituencies, how local and national inequality is changing over time, the impact of government welfare schemes, level of policing of trade mal-practices such as commodity hoarding or price manipulation, how media ownership influences its bias, whether social media helps level mass media bias or amplifies echo chambers, are aspects that if understood by citizens can help strengthen the democratic frameworks to make the world a better place for everybody. Representing the functioning of the economy in simple terms in hard – data collection from many sources, data mining to detect relationships based on state of the art in economic/financial/political models, and making it accessible to people in contextual ways – are not easy problems to solve. Advances in data science now put us in a position to make this a reality.


Our team is bringing many datasets together: (a) board of directors for all registered companies, (b) shareholder and ownership data for listed companies, (c) politicians contesting in assembly and general elections, (d) family relationships acquired from web data such as Wikipedia, (e) village level population and economic censuses, (f) district level industry surveys, (g) satellite night-lights data, (h) Landsat satellite data with image classifiers to detect built-up area, area under roads, roofing material, (i) commodity prices from agricultural mandis, (j) share prices from stock exchanges, (k) news articles from mass media sources, (l) social media pages of key people and organizations, etc. What is exciting for us is that now the technology exists to bring the data together over time, and academics and journalists have laid out what are important questions to ask from the data.



Our work is divided into three layers: infra, analysis, and applications. The infra layer includes building systems for data collection and integration, where most work so far is on building political-corporate social networks. The analysis layer involves understanding what patterns to look for in the data, drawing from financial/political/economic models created in their home disciplines; here we are analysing political-corporate interlocks, commodity time-series, mass media, census data, etc. The application layer will bring the analyses on different data sources together into concrete use-cases, and here we are building our first app to create a news feed about political-corporate interlocks. Several more ideas for apps to detect trade mal-practices and compare the effectiveness of politicians and government schemes, etc, are under development.


We started a social enterprise in 2009, Gram Vaani meaning voice-of-the-village, with the goal of building a bottom-up media platform for poor and marginalized people to voice themselves. We are now among the leading award winning technology-for-development companies with dozens of projects around the country, and a team of 50+ people working passionately towards the vision of reversing the flow of information. Our impact has been huge, but we have realized that bottom-up collective action is not alone sufficient because top-down policies and economic decisions can have far reaching effects on the lives of the poor and their efforts at uplifting themselves. GEM represents part-2 of this journey, of explaining and revealing how top-down economic processes work and affect the lives of people. Bottom-up and top-down approaches need to be bridged, by making it easier for people to understand the state of the world and then challenge the status quo.

Gram Vaani and GEM working together can redefine the way media and economies works. There is big data analysis at the quantitative census/survey population level to detect trends and surprising patterns, and there is small data analysis at the qualitative heavily sampled level to understand actual ground dynamics. The middle is missing, and we feel that we can leverage social media (or community media) data available at scale to create ethnographies which can validate patterns noticed in big data, augmented with further tools for high frequency data collection and analysis. We want to do this with economic data set against social media and mass media data, to be able to understand the economic lives of people, organizations, companies, and other connected institutions.