From the editor's desk

Focus on data

This month, a partnership between UNU-WIDER and the Groningen Growth and Development Center (GGDC) brings a new database on economic transformation public. The first findings from the Economic Transformation Database (ETD), and what they could imply, are outlined in an article that appeared in the Conversation by Gaaitzen de Vries, Emmanuel Mensah, Hagen Kruse, and Kunal Sen. The acclaimed development economist, Dani Rodrik took notice and offered his take in an editorial for Project Syndicate. The database itself contains manufacturing value-added and employment shares data for 12 sectors of 51 economies and is a significant expansion of its predecessor — the widely-used GGDC 10-sector database — updated to 2018.

In recognition of the importance of data like this to our work, the WIDERAngle is hosting a selection of articles which have a special focus on data. Collectively, they examine the potential and the limitations of different sorts of data, philosophical and methodological questions about data and its use, and what the hallmark databases UNU-WIDER keeps say about the institute, its network, its partners around the world, and our collective mission to improve societies and human well-being.

Value-neutral data?

The word data comes from a Latin root which means ‘to give’, adopted in English usage as a substitute for ‘a given’ in mathematics in the 17th century. It is no surprise, then, that research questions can sometimes be based on which answers whatever data laying around are able to give. This is in contrast to laboratory experiments, where data are collected specifically to answer the research questions we care most about.

In this respect, although data is critical to knowledge creation, it also should be suitable to the field of inquiry. When it isn't, it can mislead or misdirect. A blog by Rachel M. Gisselquist — our resident expert on inequalities between groups of people based on their identity — outlines some of the ways in which survey and census data have limitations for studying group-based inequalities. 

She asks a powerful question: what can data tell us about excluded groups when these groups are also excluded from the data?

The available official census data from India, for example, does not include information about caste, though many believe this is an important indicator of socioeconomic outcomes. A government effort to rectify this situation was undertaken with the Socio-economic and Caste Census (SECC) of 2011, but the data collected on caste have still not been released almost a decade later. This is mostly because of the enormous complexity in classifying the data — which is not just a technical problem, but a political one, too.

This illustrates a crucial point, that data is collected and stored by humans and institutions. What they choose to collect and to make available, and why, is not necessarily a value-neutral proposition.

What data does UNU-WIDER manage?

What we are entrusted to keep data on speaks to our mission. UNU-WIDER maintains hallmark databases on government revenue (the GRD) and global income inequality (the WIID), and now, structural transformation (the ETD). It was also under a UNU-WIDER project that the first complete global wealth distribution was created, demonstrating how the very existence of some data can put an issue on the map.

We are a development economics research institute, but some have called us an inequality institute because of the high profile some of our work in that area has earned. In this vein, a blog from resident data specialist, Annalena Oppel shares insights from the WIID on global trends in income inequality. But we also re-print and update the latest on the Government Revenue Dataset, from economist Kyle McNabb, and include new charts downloaded directly from the new GRD data visualization tool. Amina Ebrahim and her team provide an update on a pathbreaking project to make administrative tax data — all the data collected on firms and individuals by tax authorities — safely available to researchers for studying social and economic solutions. And, through our SOUTHMOD project, we partner with governments to develop models which can run existing national survey data through tax-benefit microsimulations to assist in calibrating policies on social transfers Our work on all these fronts is a big tell. We are interested in public policy. 

For all these examples, we compile data which is already collected. This means we are not required to grapple with the heavy ethical responsibility of choosing what data to collect, on whom, and how. In this respect we are lucky our databases can rely on the world’s source data collectors, for which we all owe a great debt.

By compiling datasets and making the data available, UNU-WIDER contributes to knowledge creation and, down the road, more well-informed policy. The hope is that the data we maintain will shed light on important aspects of development and help bring about a more equitable and inclusive world.


The views expressed in this piece are those of the author(s), and do not necessarily reflect the views of the Institute or the United Nations University, nor the programme/project donors.