Despite the fact that we are halfway through to 2030, much still needs to be done to strengthen the data and methodologies underlying the SDG indicator framework. The Resolution adopted by the UN General Assembly on 25 September 2015, Transforming Our World: The 2030 Agenda for Sustainable Development, recognized from the start the importance of closely monitoring progress on the Goals. The section dedicated to SDG monitoring notes that “our governments have the primary responsibility for follow-up and review, at the national, regional and global levels, in relation to the progress made in implementing the Goals and targets over the coming 15 years” (United Nations, 2015). The resolution also calls for broader measures of progress to complement GDP.

Thanks to the work of the United Nations Inter-Agency and Expert Group on SDG Indicators (IAEG-SDG), countries have adopted a common monitoring framework comprising 231 indicators, for which 219 have data as of October 2022.¹ Gaps remain that limit our capacity to track SDG progress, and the IAEG-SDG continues to actively expand coverage where national data is missing, developing new methods, identifying new sources to disaggregate the indicators by key population groups (such as by gender), and creating indicators to measure progress in local areas. The Cape Town Global Action Plan for Sustainable Development Data, released in South Africa at the first United Nations World Data Forum in January 2017 and adopted by the UN Statistical Commission, provides a strategic vision to strengthen data systems for Agenda 2030 (HLG-PCCB 2017).²

Since its inception in 2012, the SDSN has incorporated a strong focus on data and scienced-based pathways for sustainable development. In 2015, the SDSN Leadership Council released its report Indicators and a Monitoring Framework for the Sustainable Development Goals: Launching a Data Revolution, directed to the UN Secretary General (SDSN 2015). Through its flagship initiatives, including the SDG Index and the Thematic Research Network on Data and Statistics (TReNDS), the SDSN works closely with many partners to support global, national, and local efforts to leverage the SDGs as a monitoring and accountability tool. The SDG Index is, by design, a measure that goes beyond GDP. Building on more than ten years of work, this chapter discusses lessons learned from SDSN’s flagship initiatives on data and statistics, identifying key priorities for improving the availability, quality, and use of data for sustainable development.

Prior to the SDGs, there was already a vast body of literature on how to define and measure sustainable development (Brundtland 1987; Dasgupta and Mäler 2000; Stiglitz, Sen, and Fitoussi 2009; Arrow et al. 2013). The adoption of the SDGs and the Paris Climate Agreement in 2015 established a shared vocabulary for sustainable development, oriented towards Agenda 2030, with the Paris Agreement aiming for climate-neutrality by 2050. In comparison to previous international development agendas and goals, such as the Millennium Development Goals (MDGs), the SDGs incorporated from the start a strong focus on targets, indicators, and monitoring – notably via the annual Voluntary National Review (VNR) process. The IAEG-SDG, which operates under the United Nations Statistical Commission (UNSC) since 2015, was tasked to develop and implement the global indicator framework for the SDGs. The global indicator framework was adopted by the General Assembly on 6 July 2017 and is refined and reviewed annually. The framework currently includes 231 unique indicators (248 when including indicators that repeat under two or three different targets).

Indicators provide data in specific areas, but they do not give us an aggregate measure of a country’s SDG performance (Schmidt-Traub et al. 2017). The SDGs include 169 targets and 240 indicators, which is complex to digest from an operational point of view. Composite indices, however, despite their well-known shortcomings, allow us to synthesize complex information and may be more effective in stimulating public debate than a large number of individual scores that could result in cherry picking (OECD and JRC 2008). Widely used composite indices include the Human Development Index, the Environmental Performance Index (Wolf and Emerson et al. 2022), and the Better Life Index (OECD 2022a). The SDSN and partners have argued since 2017 that a combination of composite SDG metrics and dashboards is needed at the global, regional, and subnational levels to inform policies towards achieving complex integrated goals (Schmidt-Traub et al. 2017; Lafortune and Schmidt-Traub 2019). These metrics and dashboards can be combined with other instruments (from forward-looking models to policy trackers towards deep decarbonization and sustainable food and land systems) to increase accountability and guide action on key SDG transformations.

When it comes to integrated assessment models and tools like the SDG Index, the process often matters as much as the results. The soundness, relevance, and practical utility of such models and tools depend not only on scientific robustness, but also on their ability to generate participative exchanges, and to connect with policymakers and other stakeholders. In the context of SDG monitoring in the European Union, we proposed a framework and set of criteria in 2019, in collaboration with the Economic and Social Committee (EESC), to assess “the robustness and fitness of SDG monitoring tools” – so that they could serve as conversation-openers and contribute, along with other tools, to co-creating solutions with policymakers and stakeholders (Lafortune and Schmidt-Traub 2019).

The SDG Index measures countries’ performance on the 17 SDGs. It both tracks distance to pre-defined performance thresholds (at one point in time) and evaluates whether countries are on-track or off-track (based on past growth rates extrapolated to 2030). Building on recommendations made in Launching a Data Revolution” back in 2015, the SDG Index includes around 100 indicators (this year’s edition includes precisely 97 global indicators), clustered by SDGs and normalized on a 0–100 scale using a classic min-max function. Scores are calculated using the arithmetic mean of normalized indicators and presented for each indicator, for individual goals, and for the SDGs as a whole. Performance bounds to denote SDG achievement for individual indicators are based on a clear decision tree, similar to the one used by the OECD in its assessment of distance to SDG targets (Lafortune et al. 2020; OECD 2019a). The Dashboards address the well-known problem of “compensation” in the construction of composite indices, in which good performance on some indicators compensates for poor performance on others, by focusing on the two lowest- scoring indicators under each goal (Lafortune et al. 2018).

The SDG Index methodology is fully transparent – and available online. It has been peer-reviewed by Nature Geoscience (Schmidt-Traub et al. 2017) and by Cambridge University Press. The global edition was statistically audited by the European Commission in 2019, who recognized that, “All things considered, the SDG Index is a noteworthy effort of synthetizing the 17 adopted SDGs into a single figure. Overall, the ranks of the SDG Index are fairly robust. The index is anchored on the 2030 Agenda for Sustainable Development adopted by all UN Member States and rigorously follows the same structure of 17 goals” (Papadimitriou, Neves, and Becker 2019).

It also builds on an inclusive and participative process. The SDG Index relies on inputs from the SDSN network of experts – the largest global network of scientists and experts mobilized for the SDGs – and other partner organizations. Each year, an open online consultation is conducted using draft SDG Index results before the final report is presented. Indicator selection and performance thresholds are informed by several rounds of consultations with SDSN experts, scientists, and the general public. Around a third of the indicators come from outside official statistics (for example, from scientific papers or NGOs). This helps fill data gaps in official statistics, for instance, in the areas of international spillovers, sustainability of diets, or biodiversity. We are increasingly using space-based technologies to strengthen data availability and timeliness (boxes 4.1 and 4.2). While it takes several years to standardize international statistics, especially when methods need to be designed from scratch, our value-added is to fill existing gaps with third-party data where possible.

The SDG Index serves as a conversation opener within the research and policy community. As emphasized in the World Development Report 2021, “data alone cannot solve development problems: people … are the central actors transforming data into useful information that can improve livelihoods and lives” (World Bank 2021). We partner with regional and local organizations to prepare indices and discuss results. As an illustration, the SDG Index for Europe is prepared with and discussed among the members of the European Economic and Social Committee (EESC) – a consultative body of the European Commission that gives representatives of Europe’s socio-occupational interest groups (such as business associations, trade unions, and NGOs) and others a formal platform to express their points of view on EU issues (Lafortune et al. 2022). See Box 4.3 for additional information. Other data initiatives at SDSN, including the FABLE models for sustainable land-use systems, also rely on inclusive participatory processes, such as “scenathons” conducted by and with local country teams (FABLE 2021; Mosnier et al. 2022).

The SDG Index, including its regional and local editions, has been generally well received by the research and policy communities, and it has become the backbone of numerous collaborations with international institutions and local organizations throughout the world. Comments submitted by governments, researchers, and practitioners either publicly or privately on the global SDG Index results and methodology tend to revolve around four main perceived issues: (1) The high SDG Index scores and ranks obtained by high-income countries, including European nations; (2) Data lags, gaps, and the treatment of national estimates; (3) Questions concerning the reliability of non- official statistics and their legitimacy in the context of the SDGs; and (4) The absence of a material footprint indicator.

On point (1), our results show that rich countries generally perform poorly and are not on track to achieving environmental goals (SDGs 12–15), and that poor countries need help to combat poverty. Rich European countries top the overall SDG Index. This reflects the nature of the SDGs, as European countries, particularly the Nordic economies, perform strongly on socioeconomic goals, relatively strongly on some local environmental priorities (for example, wastewater treatment, air pollution, or deforestation), and strongly on public institutions and the rule of law. Yet the SDG Dashboards rate rich countries, including Nordic countries, at “red” on several SDGs – particularly those related to responsible consumption and production, climate action, and biodiversity – meaning major challenges remain (Lafortune, Sachs, and Schmidt- Traub 2020). Many rich countries also face a significant challenge in achieving SDG 2 (Zero Hunger), which includes unsustainable agriculture, unsustainable diets, and obesity. The stringent grading method used for the SDG Dashboards highlights negative environmental spillovers that affect climate, biodiversity, or water scarcity in other countries. Compared with other SDG monitoring reports, however, the SDG Index generates far more negative scores for rich countries on SDGs 12–15 (Lafortune et al, 2020). The most recent European edition also highlights challenges related to the “leave-no-one-behind” principle in Europe, as trends on several indicators related to inclusion are not moving in the right direction.

Some commenters have recently pointed out that a country like Bhutan, which shows remarkable commitment to sustainable development and well- being (characterized notably by its use of the Gross National Happiness Index), performs less well on the SDG Index than, for example, Finland or other Nordic countries. The rate of extreme poverty at US$2.15/day is about 6 times higher in Bhutan than in Finland, while poverty at US$3.65/day is 15 times higher; Bhutan’s maternal mortality rate (SDG target 3.1), at 60 in 100,000 live births, is more than 7 times that of Finland (8 in 100,000 live births), while its neonatal mortality rate is about 12 times that of Finland; and the incidence of tuberculosis in Bhutan is 47 times higher than in Finland. In Finland, 46 percent of parliamentarians are women, which is almost three times the rate in Bhutan, where only 17 percent of parliamentarians women. Overall, Bhutan performs lower than Finland on 14 of the 17 SDGs. The SDG Index acknowledges Bhutan’s recent progress on many socioeconomic indicators, and calls for global partnerships to promote sustainable development progress and financing globally.

On point (2), national governments often argue that the SDG Index results are biased, due to missing data and lags in data reporting. It is true that the results often represent the performance of the previous governments. There are significant time lags in international statistics, that can exceed two or even three years, as well as persisting data gaps in certain countries and country groups. This is partly due to the chronic underfinancing of statistics in LICs and LMICs.

From a methodological standpoint, we do have techniques in our methodology to address missing-data bias and time lags. Countries are included in the SDG Index ranking only if they have data for at least 80 percent of the indicators (and one criterion for indicator inclusion is that data must be available for 80 percent of countries that have at least 1 million inhabitants). Some national authorities have in the past asked to incorporate their own national estimates in the SDG Index, to address time lags and gaps. However, national estimates cannot be included in the SDG Index unless they have been submitted, approved, and published by United Nations custodian agencies or other data providers. This is essential to ensure data quality and comparability. We do include some timelier, model-based estimates (for example, for poverty or health outcomes) and geospatial data. We also review our indicator selection annually and exclude particularly outdated data points and indicators that are not frequently updated.

Strategically, the SDSN is very much committed to supporting global efforts to promote high-quality and timely data for the SDGs. Regional SDG Index editions (for Africa, Europe, and Latin America) and subnational editions (for provinces, states, regions, and municipalities) allow the indicator selection and policy discussion to be contextualized, and this data tends to be timelier (Box 4.4). In response to feedback, since 2018 we have supplemented the SDG Index with other qualitative instruments to gauge government efforts and commitment to the SDGs, in cooperation with SDSN’s global network (Sachs et al. 2022; Lafortune, Woelm, and Valentiny 2022). Finally, TReNDS and its Data For Now initiative, along with other flagship initiatives at SDSN, such as SDGs Today, aim to foster partnerships across a variety of data providers and users to unlock the potential of new technologies.

On point (3), in an effort to accurately measure often overlooked issues, such as environmental challenges and international spillovers, the SDG Index includes high-quality official and unofficial metrics that fill gaps in the official SDG metrics. For example, the SDG Index has included carbon dioxide emissions since its inception in 2016, even though a measure of greenhouse gas emissions under SDG 13 (Climate Action) was only added to the official list in 2020 (Lafortune, Sachs, and Schmidt-Traub 2020). The SDG Index and Dashboards also include unofficial measures of unsustainable fishing practices and spillovers embodied in trade and aim to incorporate more geospatial data to improve timeliness and country coverage. In most cases, these indicators went through some form of peer-reviewed process and have been published in the literature which provides some guarantees about their quality and comparability. Others are widely recognized and used measures compiled by Transparency International, the World Justice Project, and Reporter Sans Frontières (among others).

On point (4), some advocate for greater use of material footprint indicators and indicators of natural resource use (Hickel 2020). While we agree that material resource use and consumption and their impacts on the environment are important policy issues, we stand by the decision not to include indicators of material footprint or “domestic material consumption” in the SDG Index. In their current form, these indicators present well-known weaknesses. In particular, they combine by weight vastly different materials that each have different environmental impacts. Moreover, they do not correlate material flows by weight with environmental impacts, which vary tremendously across countries. For example, one kilogram of biomass used in a humid tropical country has a different footprint from the same biomass consumption in a semi-arid country. As a result, it is very difficult to compare material consumption across countries or to define targets. We recommend instead using the spillover indicators included in the SDG Index and Dashboards to capture unsustainable consumption (Lafortune, Sachs, and Schmidt-Traub 2020).

The SDG Index is a flagship instrument to promote awareness of the SDGs. The SDG Index ranking receives widespread attention from politicians and the media, which further helps to raise awareness about the SDGs and creates a “race to the top.” Increasing awareness at all levels is critical to the success of the SDGs, and the global ranking draws attention to countries’ challenges. As noted in the 2005 World Bank staff report, the main advantage of rankings is that “as in sports, once you start keeping score everyone wants to win.” However, this also creates incentives to “game the system – or corrupt it,” (Washington Post Editorial Board 2021), which is why the SDG Index is prepared by an independent group of experts and researchers, and its methodology and datasets are fully transparent.

The SDG Index is also an accountability tool that helps monitor progress and identify areas that need improvement. It is used extensively by national governments, civil society, and academia. We estimate that around 40 percent of the VNRs presented at theUnited Nations by national governments in 2021 mentioned the SDG Index. It was also listed by the European Parliament among ten composite indicators useful for policy making (EPRS 2021) and was referenced in the Parliament’s first SDG resolution in July 2022 (European Parliament 2022). Although it was not developed to be a standalone tool to inform investment decisions, the SDG Index is also increasingly used by public and private financial institutions (BPCE 2018). Alongside other data sources, it is notably being used to monitor the implementation of the first African SDG bond, issued by the government of Benin in July 2021 (SDSN 2022). See Box 4.5.

The Index also helps shed light on certain key topics, including international spillovers, and can serve as a basis to identify drivers of success and failure on SDG outcomes. For instance, our strong emphasis on quantifying domestic performance, as well as negative spillovers generated abroad via trade, has likely contributed (alongside many other initiatives) to raising awareness in the EU about such spillover effects. Our work has been referenced in policy briefs and in the literature (Arunima Malik et al. 2021; A Malik et al. 2021), a in parliament resolutions and government processes. In Europe, we partnered with Eurostat to improve the availability of data to track such spillovers, building on Multi-Regional Input-Output models, which is now a core chapter of Eurostat’s SDG report (Eurostat 2022). In addition, Box 4.6 describes how the SDG Index has been used to explore linkages between structural vulnerabilities and SDG outcomes in SIDS (in cooperation with UN Resident Coordinators in SIDS, as well as other partners), and to promote ambitious policies and financing frameworks for SDG progress (See Box 4.6).

Finally, the SDG Index contributes to global efforts to improve data availability. Over the years, countries noted that certain key data points, notably on SDG10 (Reduced Inequalities), were missing in our report. This has led some of those countries to work with the World Bank and other UN custodian agencies to compile these data. In our experience, missing data in the SDG Index is often perceived by countries as a sign of weak data capacity.

Although the SDGs have not yet completely transformed how policy is designed and implemented, as discussed in Part 3 and as is well-documented in the literature (Biermann et al. 2022; Kotzé et al. 2022; IGS 2023), they have helped to mobilize VNRs and peer learning, as well as spurring innovations in how progress is monitored, through the efforts of the IAEG-SDG. These indicators are now an important part of the evidence underpinning the more than 330 Voluntary National Reviews conducted to date to track countries’ performance towards the SDGs.

The contribution of the SDGs towards a universally accepted framework for monitoring progress is critical.

Prior to the SDGs, countries lacked a standardized method of comparing their development performance with that of their peers across a broad array of development objectives (for example, health, education, climate, ending poverty, reducing inequalities, etc.). Nor did countries have a common language to discuss and share experiences when tackling these development issues. Furthermore, the SDGs have positively impacted discourse and knowledge exchange beyond government officials. Many civil-society and private-sector actors have become SDG-conversant, facilitating greater discourse within countries across government and non-government actors.

It is difficult to assess whether the adoption of the SDGs has in itself had a positive impact on the quantity and quality of international data available for sustainable development. According to the World Bank Statistical Performance Indicators (SPI), the world progressed on average by 2.1 points over the period 2016–2022. The annual rate of progress for LICs and LMICs was faster than the world average and the HICs average, which denotes some degree of convergence. Using population-weighted averages, LMICs are now performing better than UMICs on the Statistical Performance Indicators. Progress in LMICs since 2016 has been driven by significant improvements in the SPI in some of the largest of these countries, including Bangladesh, Egypt, India, Indonesia, Nigeria, Pakistan, and the Philippines. Part of this progress might be due to investments in data capacities and statistics made during the MDG period. SIDS continue to be, by far, the group of countries with the greatest number of missing data points on the SDG Index – SIDS are missing 22 percent of SDG Index data on average, with some missing more than 50 percent.

Note: From 0 (worst) to 100 (best). Source: Authors’ calculations, based on World Bank Statistical Performance Indicators (2023).

One hypothesis, previously documented by TReNDS in the SDR 2022, is that the COVID-19 pandemic and possibly other crises may be important drivers of data innovation and collaboration. The pandemic triggered new intra-governmental collaborations to provide decision-makers with evidence to manage the crisis. Furthermore, having timely and high-quality data became a foundation for resilient and effective governments throughout the pandemic, forcing governments to adopt new processes to overcome the numerous obstacles that COVID-19 presented.

Across countries, pragmatic decisions have been made: to reprioritize staff and resources in order to modernize data capture methods and processes; to use non-traditional data sources to fill data gaps, including citizen science, social media, mobile, and satellite data; and to enhance data dissemination schemes to make it easier for policymakers and the general public to consume data. To achieve this feat, countries have embarked on a range of multi- disciplinary and cross-sector partnerships. In many countries, National Statistics Offices (NSOs), were innovators during the pandemic. They engaged in partnership activities that were previously few and far between – working with stakeholders across sectors, including civil society, the private sector, academia, and NGOs – to accelerate data innovations for policymaking and SDG attainment.

Policymakers responding to ongoing cascading crises are likely to continue the experimental and reactive approach to policy development that they adopted during the pandemic, including placing a premium on timelier and higher-quality data. As such, these crises are likely to be the primary driver of future innovations in data, to design and test public policies and programs moving forward. Thus, multilateralism and investments in global capacity-building and funding for statistics remain critical for short- and long-term improvements in information and data for sustainable development.

Building on the past ten years of work, including the SDG Index, TReNDS, and a number of SDSN’s initiatives, we draw five major lessons, which can serve as priorities to inform SDG policies and financing.

1. Science-based instruments are needed at all levels to guide SDG action and strengthen accountability. There are no magic numbers, but rather a suite of tools – including indices, integrated assessment models, policy trackers, science panels, and geospatial tools – that when combined can strengthen government capacity to implement the SDGs and to target investments. SDSN’s new flagship initiative – the SDG Transformation Center – aims precisely to provide a suite of science-based instruments and serve as a platform for peer-to-peer learning and exchange among scientists, practitioners, and investors on the next generation of SDG policy tools, analytics, and long-term pathways.

2. Additional investments are needed in capacity- building for statistics.The SDG Index and TReNDS’ initiatives have, for some years now, highlighted the acute and persisting data gaps that prevail at the global level for the SDGs, as well as the need to accelerate partnerships and investments in statistical capacity (TReNDS 2019). Although the World Bank’s Statistical Performance Indicators show signs of improvements in statistical systems since 2016, poor and vulnerable countries (including SIDS) still lack the necessary resources to implement the vision of the 2017 Cape Town Global Action Plan for Sustainable Development Data (HLG-PCCB, 2017). According to PARIS21, funding for data and statistics fell by almost US$100 million between 2019 and 2021, representing the most significant drop in funding since the start of the SDG era (OECD 2022b). And as highlighted at the UN World Data Forum in April 2023 and reiterated in the 27 April Hangzhou declaration: statistical capacity in the poorer and most vulnerable countries requires “an urgent and sustained increase in the level and scale of investments in data and statistics from domestic and international actors, from the public, private, and philanthropic sectors” (HLG-PCCB 2023).

3. We need to invest in data and science literacy to strengthen the science-policy interface. According to major international studies, few 15-year-old students can tell the difference between a fact and an opinion (OECD 2019b). In an information-rich and post-truth environment, citizens and decision- makers need knowledge and tools to transform data and science into evidence, actions, and long-term policies. Yet UNESCO estimates that there is a nearly US$100 billion finance gap for countries to reach their education targets (UNESCO 2023). The SDSN and its partners are increasingly collaborating with governments and parliaments – and also working closely with business associations, private financial institutions, trade unions, and academic organizations – to improve data literacy and support science-based policy discussions to advance the SDGs at the national and local levels. Strengthening the science-policy interface is key for implementing long-term pathways for sustainable development.

4. Non-traditional statistics and science-based pathways help to address shortfalls in official statistics; they could be further leveraged to inform investment decisions. Developing official international statistics takes time. Especially when no suitable methodologies or data-collection methods exist. Halfway into the SDGs, for instance, we lack a good-quality international metric, available to most countries, to track “mechanisms in place to enhance policy coherence of sustainable development” – indicator 17.14.1. Nevertheless, building on the improvements in the field of industrial ecology, the SDSN and its partners have included an evaluation of negative international spillovers (one component of sustainable development policy coherence) in the SDG Index since 2017. Non-official sources of statistics, such as citizen science, social media, earth observation data, artificial intelligence (AI), model-based estimates, and other pathways produced by academics and researchers, support greater accountability and can provide a forward-looking evaluation of efforts to implement the SDGs, complementing official statistics. Additionally, initiatives like the Climate Action Tracker help provide science-based and forward-looking assessments of countries’ ambitions and actions taken to further key SDG transformations. These types of assessments, as well as science-based decarbonization targets and food and land pathways, can be further leveraged in the design and assessment of public and private investment programs for the SDGs, including sustainability-themed bonds.

5. Space-based technologies help address data gaps and timeliness, including supporting the “leave no one behind” principle; they can be further leveraged via global partnerships. Time lags in international data reporting can exceed two to three years, including for key SDG indicators. With the elevated focus on and interest in data, COVID- 19 has set the stage for new user expectations, with many users – especially the general public – now expecting to obtain data in real time (Sachs et al. 2022). The global community has mobilized space-based technologies to help provide more timely and granular information on the state of the global commons, or on access to key services (among others). In Europe, for instance, Copernicus, the EU’s flagship Earth Observation and Monitoring program, was mobilized early on to improve SDG data availability, timeliness, and granularity. And the new SDGs-EYES program will establish “an integrated scientific, technological and user engagement framework, overcoming the knowledge and technical barriers that prevent the exploitation, combination and cross-feeding of data and tools” to support SDG action. At the global level, UN-GGIM and the Group on Earth Observations (GEO) aim to reduce the technical and legal barriers to using geospatial data and to strengthen multistakeholder partnerships. And TReNDS’ Data for Now initiative is working to support countries’ capacity to deliver robust and timely data to achieve the SDGs through user- centric approach, multistakeholder partnerships, and use of alternative data sources, including space-based technologies. The “TReNDS Data for Now” initiative draws on a user-centric approach, multistakeholder partnerships, and the use of alternative data sources such as space-based technologies to build countries’ capacities to deliver accurate and timely data to achieve the SDGs.