Why is there so little data on who faces poverty and where they live?
Everyone seems to want data or more data. Data that helps target the 'right' people with the 'right' things. Data on the billions who suffer deprivations as legitimation for the entire aid and development business. Data to measure and monitor the global goals and targets that governments have agreed to (or hopefully will agree to within the proposed Sustainable Development Goals (SDGs).
There is also an enthusiasm for mining existing datasets. All this was highlighted by A World That Counts, a report prepared at the request of the UN Secretary General. This suggests that the data revolution can be a revolution for equality.
Data for whom?
But this demand for data generally comes from those who control funds and other resources. National governments and international agencies. Not those who need funds (or schools or potable water…).
Surely the point of data in development is to make sure it assists (and even empowers) those in need to get what they lack. It is a means of informing those with the powers and responsibilities about these needs. But most of the official data sets used to assess who has unmet needs cannot tell us who these people are or where they live.
These official data sets are drawn from national sample surveys. Their sample sizes are too small to show who within a nation's urban (or rural) population is facing deprivation, the nature of their needs or where they can be found. So they provide no useful data to local governments, even though so many of the existing Millennium Development Goals (MDGs) and proposed SDGs depend on local government for their fulfilment.
This can result in a misleading, inaccurate and incomplete picture. Much of the data being used to monitor the MDGs, for example, is inaccurate because it does not measure the goal – for instance the indicators for water provision do not assess whether the water is safe to drink or the water supply is reliable or accessible.
Much of it is also incomplete because what is reported is often based on guesses or projections since no country gathers data each year on all the MDG goals and targets.
Oddly enough, the one data source that can provide detailed local data – down to the particular needs in each street – are censuses, but these are rarely mentioned. They are not even considered in A World that Counts. They may seem expensive yet they can provide very valuable data to local governments, local civil society organisations and other local agencies about needs in each street or ward.
Local, local, local
Almost all development is 'local' in the sense of the needs (including data needs), the actions to address them and the much needed accountability (to local populations). So why do we focus mainly on generating data at a national level by national or international agencies, data that does not contain the detail needed to guide local action? Why does A World that Counts focus so much on strengthening national agencies without mentioning how these should be working with local governments to meet local data needs?
In contrast, local communities have shown that they can collect relevant, informative and significant data – as documented in a remarkable report by the Asian Coalition for Housing Rights. In six Asian nations, elected representatives of the urban poor are showing their national statistical offices how best to define, measure and monitor urban poverty.
So what other data sources are useful for local decisions and actions? In urban areas, perhaps the most powerful are the surveys of all 'slums/shacks' in a city done by their inhabitants with the support of slum/shack dweller federations.
Around a billion people live in informal settlements in urban areas. Surveys in Nairobi, Harare and Cuttack (PDF), for example, have provided detailed data on each informal settlement. These surveys can be the equivalent of censuses with data collected from each household (and returned to that household and to community organisations for checking). These provide the data needed for local discussions of priorities and guides to where development investments are needed.
As the information is owned by the local inhabitants and their organisations, it also helps ensure that their needs and priorities are fully represented. But these get no mention in A World that Counts.
Local democracy as a data generator
Where local democracy works well, democratic processes can replace data. Instead of data gathered on (say) who has inadequate provision for water, sanitation, drainage, health care…, local governments can find out about these needs through democratic processes.
This becomes particularly powerful when local governments work with grassroots organisations and federations formed by residents of informal settlements. This can then be enhanced greatly by participatory budgeting – where the needs for funds are defined, prioritised and acted on by representatives elected in each district of a city. Also where all details of local government revenues and expenditures are made public.
Both the data generated by the slum surveys and the discussions around resource allocations for participatory budgeting are intensely local. Unlike national sample surveys. It is also worth recalling that so much 'good' development in what are today's high-income nations was a response by local governments to democratic pressures.
The current response to the lack of data about poverty is heavy investment in more data gathering by national governments and international agencies. But this still fails to address the data needs of local governments and civil society organisations.
What we need is far more attention on how the needs and priorities of those facing deprivations are recorded and used in each locality (in ways that are accountable to them).
Every initiative to improve data collection or mine existing datasets has to ask – will this benefit those with unmet needs? Will it increase their influence – and their capacity to act and to hold government and international agencies to account? There is a lot of excitement about mining data from new sources – for instance mobile phones. But you cannot eat data. Or a mobile phone.