Despite all the fanfare of the launch of the SDGs in 2015, and the widespread call for a “data revolution,” national governments and the global funding community have not come forth with the scale of funding needed for data to make and measure progress on the SDGs.
The Bern dialogue has an opportunity to start a new and better conversation about financing data for the SDGs. It can overcome three shortcomings from previous discussions. The first shortcoming has been a focus on bottom-up costing exercises (see examples by SDSN, Paris21, and World Bank and WHO), which tally the cost of filling all relevant data and capacity gaps, but stop short of establishing priorities or considering potential efficiencies. The second shortcoming is a narrow focus on official statistics, which are deeply central to but are not solely responsible for meeting data needs for the SDGs. The third shortcoming is a single-minded focus on data production with little attention to increasing capacity and incentives for data to be used for government decisions.
Participants in Bern and others advocating for data financing can do two things differently: think like a data user, and take a portfolio approach to funding data.
1. Think like a data user
In 2017 the World Bank evaluated more than 200 statistical capacity building projects and found only 27 projects supported capacity building for data users. The evaluation calls for the Bank to foster a “user-centered data culture,” yet finds that despite decades of investing in capacity for data production, the Bank has “not yet formulated a conceptual model for assessing (or presumably supporting) user capacity.” So what might data funders do better going forward?
At the very least, financing for data production should require (and provide funding for) deep engagement with the expected users from the outset. A user-centered data culture also has to focus on officials’ motivations to invest in and use data. Many policymakers—including the ministries of finance that we hope will allocate domestic resources for statistics—are human beings first and foremost. Most human beings (except us few data evangelists) are not personally moved by a long list of data gaps. But they can be deeply moved by stories of how data help solve problems they care about. So, when you are engaging top officials who set budgets in your own institutions and in partner countries (since we expect much of new funding to come from country governments themselves), take some stories with you. Take a look at the Data Impact stories from Open Data Watch, gender data impact stories from Data2X, and ten stories on the returns on investing in data from the Global Partnership for Sustainable Development Data (GPSDD) and SDSN TReNDS. Also share examples of burning policy priorities that can’t be met simply because we don’t have the data necessary to act. These stories can help you make the case for more data finance and for a focus on data use. So with this ammunition in your pocket, friends in Bern please carve out some time to discuss what more you can do (and fund) to increase government incentives and capacity to use data.
If you need a little inspiration, take a look at the UK Department for International Development’s Building Capacity to Use Evidence program, for which an excellent evaluation is available, and its forthcoming successor program, Strengthening the Use of Evidence for Development Impact. To create incentives for data and evidence use, the Mexican national council for evaluation of social development policy, CONEVAL, presents awards to ministry officials who champion the use of evaluation findings to improve government. As the hard-working champions within the Ghana Statistical Service often say, they want to see their data used by government colleagues. We mustn’t take that side of the equation for granted.
2. Take a portfolio approach to investments in data
A portfolio approach has two key ingredients: diversity and efficiency.
Just like any other investment portfolio, investments in data for the SDGs require a diversity of instruments, balanced across financing tools, timelines, sectors and risk tolerance. The World Bank or major bilateral donors might support long-term, multi-country capacity-building efforts while private sector firms offer data philanthropy or lend data analytics capacity. Foundations might be best positioned to support non-governmental organizations, fund research into new data methods, facilitate cross-country learning, or take risks on new partnerships. Likewise, meeting and measuring the SDGs will require a mix of data types, from censuses and surveys to administrative data, and from small area estimates to satellite imagery. Hopefully whatever strategy comes out of the Bern meeting will reflect and promote this diversity.
A portfolio approach starts with goals of what value one wants to create (in this case, using data to help achieve the SDGs) and then considers the most efficient mix of investments to get that value. A strong portfolio needs to optimize across a set of investments—filling data gaps, increasing capacity and incentives for data use, addressing challenges like interoperability to get maximum value from each data product (with help from this practitioner’s guide from GPSDD and the UN Statistical Division), and overcoming policy and legal barriers. On the policy side, national statistical offices (NSOs) are being asked to innovate, including incorporating private sector data that comes with privacy and legal concerns that few NSOs are equipped to manage. That’s why the first recommendation of the 2017 OECD Development Cooperation Report is to “make statistical laws, regulations and standards fit for evolving data needs.” It’s far more efficient to address these challenges collectively than leave each country to struggle through on its own.
A key part of efficiency is ensuring that investments don’t work at cross-purposes. Shockingly, according to analysis by Development Gateway, funders spend more on hindering country data systems than supporting them. As of 2016 approximately 2-3% ($2.85-$4.28 billion) of overall development assistance went towards project monitoring and evaluation (M&E). This M&E is a well-intentioned focus on results, but mostly consists of collecting project-specific data that are rarely used or shared, and for which the burden of data collection often falls upon local providers at the expense of their own work. It’s not just that every dollar spent on collecting project data is a dollar not spent on systems improvements; it’s also a distraction that can undermine statistical system development. Data investment portfolios need to be scrubbed of distractions and distortions.
So, friends meeting in Bern, please come together to think like a data user and apply a portfolio approach. Meeting the SDGs depends on all of us to do so.