This meeting, held on 28 June, was also a webinar to allow for members overseas (or not in London) to attend and contribute and was filmed for the RSS YouTube channel. Twenty members were present and a further seven joined the webinar.
Ian Plewis, chair of the International Development Section introduced the main contributors Mary Strode (independent statistical consultant), Johannes Jütting (manager of PARIS21 – Partnership in Statistics for Development in the 21st Century), Claire Melamed (executive director of the Global Partnership for Sustainable Development data), Neil Jackson (chief statistician, UK Department for International Development) and Michael Woolcock (lead social scientist, World Bank Development Data Group).
Developing the capacity of statistical systems to monitor progress regarding the Sustainable Development Goals (SDGs), cuts across the work of government, academia, the private sector, NGOs and individuals. At the heart are the national statistical systems (NSSs), centred on the national statistical offices (NSOs), who in many cases still face severe challenges linked to human, physical and financial resources, combined with issues of governance and status.
Drawing on some decades of experience with Statistical Capacity Development (SCD – previously SCB – statistical capacity building), Mary Strode started by asking the question – Are we winning? SCD projects have been running for years but the World Bank Statistical Capacity Indicator shows almost no improvement between 2005 and 2017. Admittedly this is a very blunt measure – no doubt there are a lot more data being collected and disseminated in the public domain, but does this reflect an improvement in capacity? The needs of international agencies, backed by donor money, have unbalanced the work of NSOs in favour of fully-funded surveys with international comparability – not in themselves a bad thing, but leaving certain aspects of statistics very much on the sidelines. The advent of the Millennium Development Goals focussed efforts still further, and the SDGs are carrying this on.
Mary sees three challenges for SCD:
- A political enabling environment for statistics is missing in many countries
- Statistics has become detached from national governments and is instead an extractive industry for the international community – this needs to be reversed
- Many think that IT will solve the problem – but this is not the only aspect of capacity building which needs attention.
She sees three long-held assumptions that have hindered progress:
- The job of statisticians is to satisfy user demands, but we have not said which users and often draw the user boundary at the edge of the NSS
- Weak planning is the problem, since people in general do not know how expensive it is to collect data – but governments who don’t use or value data are unlikely to fund it;
- Stronger regulation is a solution, but this does not work in a weak environment.
Johannes Jütting took a less gloomy view about what has been achieved, but agreed that there was still much to do. Since its formation, PARIS21 has supported statistical systems around the world with tools, information and advocacy, and has seen the debate around capacity development broaden. At the World Data Forum (Cape Town, January 2017) a reflection on capacity development was started, and one of the outcomes has been PARIS21’s CD4.0, which he then presented.
Statistical Capacity can be defined as 'the ability of a country’s national statistical system, its organisations and individuals to collect, produce, analyse and disseminate high quality and reliable statistics and data to meet users’ needs'. The CD 4.0 Framework identifies three levels at which capacity development can enter: individual, organisational and system; and five targets: resources, skills and knowledge, management, politics and power, and incentives, giving 15 categories of capacity development. For example Organisational + Resources would cover human resources, budget and infrastructure. System + Management would cover NSS coordination mechanisms, coordination of the data eco-system and the advocacy strategy. Individual + Politics and Power would cover teamwork and collaboration, communication and negotiation, and strategic networking.
Why is it necessary to have these guidelines? Simply, it has been accepted that there is a wider range of capacities which need to be developed beyond the skills within the national statistical office, and the concept of what capacity means has been broadened beyond statistical skill, for example taking in aspects of management. It is worrying that at the moment we seem to be going back to donor-specific data for monitoring and evaluation and away from building systems. The guidelines are for beneficiaries and donors.
Johannes finished by asking three questions:
- Do guidelines matter?
- Does CD 4.0 get it roughly right?
- Do we also need capacity development for non-government actors, particularly civil society organisations?
Clare Melamed began the discussion by reflecting that although it was critical to focus on capacity development, it would not solve all of our problems if we only considered the supply side. Use of data for decision making must be part of the evaluation of capacity. But it was pointless to talk to users about data: what is needed is to talk to them about what they can use data for, to find out what makes people want to 'cross the data boundary'.
Statistics have to be disseminated in the way people want to receive them, implying improved communication skills on the supply side. The NSOs have lost the monopoly on data production, so partnerships are now required, with the leadership and incentives to make them work. Capacity development has to be done in the context of rapid change. And finally, are we developing the capacity of the data eco-system of today, or of the one we want for the future?
Neil Jackson continued the discussion by saying that the transition from the Millennium Development Goals to the SDGs has helped to open up country ownership of national data and to broaden the political interest in what is published, particularly when inter-country comparisons are made public. And from political interest to user needs to a recognition that national capacity needs developing is a chain which should be encouraged. We improve through leadership, through tools and processes, and through skills development.
Neil added that for many countries (including the UK) populating the SDG indicator set is a daunting prospect, and donors need to help countries to set their own priorities for a key SDG sub-set rather than impose one.
Michael Woolcock continued this theme by asking how donors work out which is the most locally appropriate and practicable entry point for a capacity development programme. The temptation is to offer what we know and have already developed, but that was the way of 'best practices' which he felt was not effective. We must get away from offering solutions and move into helping people to analyse their own priorities. For this, the CD 4.0 grid could be a very useful identification tool.
It is undeniable that developing country governments need help. The international community has placed enormous pressures on them to deliver, and increasingly their people are doing the same, when they begin to realise what they can expect.
In the general discussion which followed, a number of points were made:
- The role for academic institutions – expert analysis has a key role to play, but we cannot assume that there are sufficient expert users out there – they need regular training
- The need to get data out of the statistics offices and into the public domain in the form of open data, not just the headline figures
- Sub-national data can often be the key to increasing demand
- As can peer reviews by neighbouring statistics offices.