How Much Will a Data Revolution in Development Cost?

Abstract

The UN High Level Panel has called for a data revolution. The world’s population should be counted, measured, weighed, and evaluated. This information should be collected, compiled, aggregated, and presented in such a form that it can usefully inform policy-makers and citizens in aggregated forms, and disaggregated according to region, village, gender, and population group. It is tempting to think that having more and better information will improve policy choices, but this is not automatic. Moreover, demanding more information may lead to a deteriorating supply of information if demand outstrips supply. This article therefore focuses on the cost of the data revolution rather than its possible benefits. The simple starting point is that data do have a cost. The cost of the data revolution will be considerable. The costs include the sheer financial cost of monitoring, but importantly extend to include the opportunity cost of competing demands on survey capacity. Particular indicators also influence behavior by skewing activities in favor of completing goals with quantifiable targets. More data are only better data if they contain meaningful information that clearly outweighs the opportunity cost related to its gathering. However, funding available for sustainable development goal (SDG) measurement is limited and there are also institutional capacity limits to be reckoned with. On 11 March 2016, the United Nations Statistical Commission’s Interagency and Expert Group on SDG Indicators (IAEG-SDGs) agreed on 230 individual indicators to monitor the 17 goals and 169 targets. Because the final list of 230 indicators has been a moving target, the research for this article took as the starting point of what a reasonable level of measurement of the millennium development goals would have cost to raise a debate and to give a crude estimate over the cost of the introduction of the SDGs, with their exponential increased demand for data. This is done by bringing together and reconciles the existing information on the costs measurement as a reference for scholars concerned with SDG feasibility and operationalization.

Keywords:

• sustainable development goals
• statistical capacity
• population census
• poverty survey
• United Nations

Notes on contributor

Morten Jerven is Professor in Development Studies. He has a Ph.D. in Economic History from the London School of Economics and has published widely on African economic development and particularly on how we measure development. He is the author of Poor Numbers: How We are Misled by African Development Statistics and What to do About it. Ithaca, NY: Cornell University Press, 2013; Economic Growth and Measurement Reconsidered in Botswana, Kenya, Tanzania, and Zambia, 1965–1995, Oxford University Press, 2014; Africa: Why Economists Got it Wrong, New York: Zed Books, 2015; Measuring African Development: Past and Present, Routledge, 2015, and Statistical Tragedy in Africa? (with Deborah Johnston), Routledge, 2015.

Supplemental data

Supplemental data for this article can be accessed at http://dx.doi.org/10.1080/08039410.2016.1260050

Notes

1 In recent years, Nigeria has received about 2 billion USD in Official Development Assistance, according to OECD Development Assistance statistics.

2 Virtual Statistical System, (2014).

3 Later expanded to 60.

4 Though not strictly measuring the survey cost of the MDGs, in 2004, the Marrakech Action Plan for Statistics, an initiative of the World Bank, regional development banks, and the Organization for Economic Cooperation and Development (OECD), estimated that the annual cost of improving both national and international statistical systems up to acceptable levels would be somewhere between US$140 and 160 million per year in additional resources.

5 Despite declarations being made about MDG1 (halving the ratio of poor who live in extreme poverty) being met in 2015. The data are not likely to be ready until sometime in 2019; currently, we have some data for some countries for 2012.

6 This is a revision of Jerven (2014c), reflecting some changes that have been announced since then.

7 It was not possible to obtain cost information about MICS, so these were excluded in the costing of the MDGs.

8 In addition, I wanted to add an annual MICS but, as discussed below, cost data were not available on this survey instrument, and it was therefore not included in the resulting analysis, nor the final estimate.

9 For example, a country with rugged terrain, low literacy, or weak infrastructure would be much more difficult to survey.

10 See the Appendix in Sustainable Development Solutions Network, 2015 for details.

11 As per email with Susan McInturff and Martin Vaessen of Measure DHS, 15 January and 27 January 2014, respectively.

12 Although the World Bank CWIQ handbook listed a 10,000 household study and $330,000 yearly cost, I found household sample averages for medium- and large-sized population countries to be higher, and have adjusted our costs accordingly to better represent the reality of the CWIQ surveys previously taken.

13 As per email correspondence and telephonic conversation with Tara Moayed of UNICEF, 30 January 2014.

14 1bn per year is 15 bn times 4 or 10. These dollar amount are not corrected for inflation in measurement costs.

15 See discussion in Jerven (2015).