The true difficulties posed by SDG measurement

Publication Date: 

Wednesday, 7 January 2015 - 12:27pm


Measure What Matters

Prof. Bjørn Lomborg, founder of the Copenhagen Consensus Center and author of The Skeptical Environmentalist, sets out the real challenge of monitoring the SDGs in a cost-effective way. If current development and data collection budgets will struggle to cope with the proposed raft of targets and metrics, can SDG measurement really succeed without better alignment of existing reporting frameworks?

At the start of the twenty-first century, the international community made some smart and simple promises with the so-called Millennium Development Goals. The world would halve the proportion of people suffering from hunger and living in extreme poverty, achieve universal primary education, and dramatically reduce child mortality by 2015. There have been many successes, though not all of the MDGs’ targets will be achieved.

The target of halving hunger, for example, may be missed – though not by much. In 1991, 23.4% of all people in the developing world were malnourished; more than a billion people went to bed hungry. By 2013, the proportion had dropped to 13.5%. Though the developing world had 1.7 billion more people than in 1991, 209 million fewer were starving. Over the past 22 years, the world has managed to feed almost two billion more people adequately – no small feat.

Over the next year, the world’s 193 governments will come together to set new global targets to be met by 2030. The task amounts to this generation’s greatest opportunity to translate high aspirations into concrete targets. But choosing the targets that will do the most good requires learning from current experience.

That choice should be based on a comparison of the proposed targets’ economic, social, and environmental costs and benefits. My think tank, the Copenhagen Consensus, has asked 60 teams of the world’s top economists to examine most of them.

Of course, high-profile issues that affect people’s daily lives – for example, health, education, food security, clean water and sanitation, and the environment – attract the most attention. But we cannot simply assume that efforts to improve the situation are effective; we also need to measure how well we address these issues, and measurement has real costs. After all, money spent on one priority is not available for others.

In a recent paper for the Copenhagen Consensus, Morten Jerven of Simon Fraser University examines how much measurement will cost – and how much the international community can justify spending. And measuring turns out to be much harder than one might expect.

For many indicators in the developing world, there is only a small amount of information about what has been achieved. There is no shortage of information available online about the number of poor people in almost any country in any year since 1990; but much of it is based on sketchy data.

To estimate the number of poor in a country requires a household survey of consumption. But six of the 49 countries in Sub-Saharan Africa have never had a household survey, and only 28 countries have had one in the past seven years. For example, according to the World Bank, 11.92% of Botswana’s population was poor in 2008. But these data are based on just one household survey – from 1993.

Actually, there have been very few recent surveys, and most of the available numbers are projections and estimates, not hard data. Overall, there are more gaps than real observations, and the observations themselves are often dubious.

Data collection for the MDGs was patchy, and the quality of much of the information collected was questionable. After assembling available information about survey costs around the world, Jerven estimates that proper monitoring of all 18 targets and 48 indicators would have cost $27 billion. This sounds like a lot, but it is only 1.4% of the roughly $1.9 trillion spent on development aid during this period.

In 2013, a high-level panel of politicians and civil-society and private-sector leaders advocated the creation of “better data-collection systems, especially in developing countries.” Likewise, the so-called Open Working Group called on the world to “increase significantly the availability of high-quality, timely, and reliable data disaggregated by income, gender, age, race, ethnicity, migratory status, disability, geographic location and other characteristics relevant in national contexts.” And last month, UN Secretary-General Ban Ki-moon specifically proposed establishing a “comprehensive program of action on data.”

The problem is that the next set of targets is growing ever larger. The high-level panel suggested 59, compared to 18 under the MDGs, and the Open Working Group nearly tripled the total number again, to 169 targets. Jerven estimates that carrying out even minimal data collection for all 169 would cost at least $254 billion – almost twice the entire annual global development budget.

Even this is a low estimate. For starters, it does not includes the costs of national governments’ gathering of basic administrative data, or of all of the recommended household surveys, because these costs were impossible to obtain. And gathering data in countries where none has yet been collected will likely prove even costlier.

Moreover, there is no allowance for maintaining national statistical offices, training and retaining personnel, or analyzing and disseminating the data. Given capacity constraints, particularly in the poor countries, this cost is likely to be high. Indeed, analyzing more long-term data for the international donor community may impede finance ministries’ ability to make high-frequency data available for running the country. The MDG agenda has already stretched statistical capacity; 169 new targets will only make the situation worse.

The MDGs were effective in part because the number of targets was limited. So it is reasonable to consider how many targets we could measure properly, rather than how to collect data on a large number of them.

A useful point of comparison is what industrialized countries spend on their statistical services. For example, the Norwegian and British governments spend about 0.2% of GDP. Using this figure as a rough indicator of willingness to pay for proper measurement and monitoring suggests that the post-2015 goals should include just four targets.

As the international community gears up for the next development agenda, the danger is that we let our ambition overpower our reason. If we embrace too many targets – and especially an unwieldy 169 – we risk doing a disservice to the world’s neediest people.

This article has been reposted in-part from the Project Syndicate website, and is available here in full.


Add new comment

This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.
Target Image