2. Leaving No One Behind

The Sustainable Development Goals represent humanity’s shared commitment to achieve ambitious global gains for people and the planet by 2030. Of the SDG’s 17 goals and 169 targets, not a single one is detached from the implications and potential of digital technology. From ending extreme poverty, to promoting inclusive economic growth and decent work, to reducing maternal mortality, to achieving universal literacy and numeracy and doubling the productivity of small farmers – progress is intertwined with the use of digital technology and new forms of digital cooperation.11 

However, technological solutions are not enough. Diverse political systems, history, culture, resource constraints and other factors which have marginalised far too many people, are – and will continue to be – of critical importance. The application of technology must be aligned with investments in human capital, infrastructure and environmental protection. Widening access to digital technologies is necessary, but not sufficient. Access needs to be affordable to be meaningful. Special efforts are needed to remove barriers for marginalised groups who often face a double bind: they already face discrimination in its many analogue forms and are least likely to be connected. Pre-existing forms of marginalisation should not be perpetuated or aggravated in the digital sphere. 

Success will require a commitment by all involved stakeholders to hard work and learning over many years about how to broaden opportunity and build truly inclusive economies and societies. We believe that there is significant room for digital technology and improved cooperation to contribute to these efforts.


With mobile internet and increasingly powerful and lower cost computing, every person can theoretically connect to anyone else, obtain and generate knowledge, or engage in commercial or social activity.12 For organisations of whatever size, likewise, there are fewer technical barriers to global economic interaction at scale. Digital technology can support economic inclusion by breaking down barriers to information, broadening access, and lowering the level of skills needed to participate in the economy.13 

Of course, this does not mean that everyone and everything should be connected or digitised. Nor does it mean that the social and economic consequences of digital technology are necessarily inclusive or beneficial. Digital technology can both provide opportunity and accentuate inequality. 

The challenge for policy makers, and other stakeholders seeking to contribute to progress toward the SDGs, is how to cooperate to leverage technology to create a more inclusive society. As we emphasise in this chapter and our recommendations, we believe digital cooperation must steer how digital technologies are developed and deployed to create meaningful economic opportunities for all. 

Developing an inclusive digital economy will require sustained and coherent effort from many stakeholders across all walks of life. National policy frameworks and international agreements need to find ways to promote financial inclusion, innovation, investment and growth while protecting people and the environment, keeping competition fair and the tax base sustainable. 

Developing an inclusive digital economy will require sustained and coherent effort from many stakeholders across all walks of life. National policy frameworks and international agreements need to find ways to promote financial inclusion, innovation, investment and growth while protecting people and the environment, keeping competition fair and the tax base sustainable.


The ability of digital technologies to empower traditionally marginalised people and drive inclusive economic development is illustrated by financial inclusion.14 Mobile money, digital identification and e-commerce have given many more people the ability to save and transact securely without needing cash, insure against risks, borrow to grow their businesses and reach new markets. 

According to the World Bank’s Global Findex 2017 report, 69 percent of adults have an account with a financial institution, up by seven percentage points since 2014. That means over half a billion adults gained access to financial tools in three years. But many are still left behind, and there is scope for further rapid progress: a billion people who still have no access to financial services already have a mobile phone. 15 

Mobile money – the ability to send, receive and store money using a mobile phone – has brought financial services to people who have long been ignored by traditional banks.16 It reaches remote regions without physical bank branches. It can also help women access financial services – an important aspect of equality, given that in many countries women are less likely than men to have a bank account. 

New business models enable people who have no physical collateral to demonstrate to lenders that they are creditworthy – for example, by allowing the lenders to see phone location data and online transaction and payment history.18 Mobile finance matters in wealthy countries, too, where low-income and historically marginalised groups generally both pay higher interest rates and receive a narrower range of financial services.19 

Well-known examples of mobile money include Kenya’s M-Pesa and China’s Alipay. Launched in 2007 by Vodafone, M-Pesa received support from diverse stakeholders who all have a role to play in digital cooperation. A private sector innovation with donor funding, it originally addressed microfinance clients in partnership with civil society – then citizens found new uses, including low cost person-to-person transfers.20 Alipay has made millions of small business loans to online merchants, more than half of whom are aged under 30. 21 

What works in one country may not work in another.22 Rather than try to replicate specific successes, digital cooperation should aim to highlight best practices, standards and principles that can create conditions for local innovations to emerge and grow based on local issues, needs and cultural values. India, for example, has added 300 million bank accounts in three years as new business models have been built on the India Stack, a set of government-managed online standards in areas including online payments and digital identity. 23 

Across many areas of financial inclusion, fragmented systems and lack of cooperation within and across states make it difficult to fully realise the benefits of digital technology. Common standards for cross-border interoperability of mobile money could unleash much more innovation: discussions to develop them should be a priority for digital cooperation. 24

Digital identification (ID) can support inclusive economic development more broadly. More than a billion people today lack an official way to prove their identity: this means they may not be able to vote, open a bank account, transact online, own land, start a business, connect to utilities or access public services such as health care or education.25 The consulting firm McKinsey & Company studied seven large countries and concluded that digital ID systems could add between 3 and 13% to their gross domestic product.26 

However, digital ID systems require caution. A digital ID can help unlock new opportunities but can also introduce new risks and challenges. They can be used to undermine human rights – for example, by enabling civil society to be targeted, or selected groups to be excluded from social benefits.27 Data breaches can invade the privacy of millions. To minimise risks, countries should introduce a digital ID system only after a broad national conversation and allow for voluntary enrolment and viable alternatives for those who opt out. They should establish ways to monitor use and redress misuse. Countries could cooperate to share experience and best practices in this regard. 

The World Bank Identification for Development (ID4D) initiative has identified ten Principles of Digital Identification covering inclusion, design and governance “to improve development outcomes while maintaining trust and privacy”.28 This initiative draws on the experiences of countries that have already implemented digital ID systems. Among the most successful is Estonia, where citizens can use their digital ID to access over 2,000 online government services. Building on the positive and cautionary lessons of early adopters, the Modular Open Source Identity Platform (MOSIP) is developing open source code countries can adapt to design their own systems.29 

Recent years have also seen a dramatic increase in e-commerce, including by individuals and small businesses selling products and services using online platforms. When e-commerce platforms provide technological services to small entrepreneurs, rather than compete with them, they can level the playing field: it is relatively cheap and simple to start a business online, and entrepreneurs can reach markets far beyond their local area. 

Inclusive e-commerce, which promotes participation of small firms in the digital economy, is particularly important for the SDGs as it can create new opportunities for traditionally excluded groups. In China, for example, an estimated 10 million small and medium-sized enterprises (SMEs) sell on the Taobao platform; nearly half of the entrepreneurs on the platform are women, and more than 160,000 are people with disabilities.30 E-commerce can support rural economic inclusion as clusters of villages can develop market niches in certain types of products: in China, an estimated 3,000 “Taobao villages” have annual online sales of more than one million dollars annually.31 A growing e-commerce sector also creates demand and employment in related businesses including logistics, software, customised manufacturing and content production. 

E-commerce shows how digital technologies with supportive policies can contribute to inclusive economic development – it has done best in countries where it is relatively easy to set up a business, and where traditionally neglected populations are able to get online.32 As with inclusive mobile finance, as more individuals and small enterprises buy and sell internationally, there is also a need to create more supportive rules for cross-border e-commerce. 

As e-commerce grows, there are also concerns about its relation to local and international markets, as discussed below in Section 2.3. 


The immense power and value of data in the modern economy can and must be harnessed to meet the SDGs, but this will require new models of collaboration. 

The Panel discussed potential pooling of data in areas such as health, agriculture and the environment to enable scientists and thought leaders to use data and artificial intelligence to better understand issues and find new ways to make progress on the SDGs. Such data commons would require criteria for establishing relevance to the SDGs, standards for interoperability, rules on access and safeguards to ensure privacy and security. 

The immense power and value of data in the modern economy can and must be harnessed to meet the SDGs, but this will require new models of collaboration.

We also need to generate more data relevant to the SDGs. In a world which has seen exponential growth of data in recent years,33 many people remain invisible. For example, the 2018 UN SDG Report notes that only 73 percent of children under the age of 5 have had their births registered.34 The World Health Organization (WHO) estimated in 2014 that two-thirds of deaths are not registered.35 Only 11 countries in sub-Saharan Africa have data on poverty from surveys conducted after 2015. Most countries do not collect sex-disaggregated data on internet access. 36

Anonymised data – information that is rendered anonymous in such a way that the data subject is not or no longer identifiable – about progress toward the SDGs is generally less sensitive and controversial than the use of personal data of the kind companies such as Facebook, Twitter or Google may collect to drive their business models, or facial and gait data that could be used for surveillance.37 However, personal data can also serve development goals, if handled with proper oversight to ensure its security and privacy. 

For example, individual health data is extremely sensitive – but many people’s health data, taken together, can allow researchers to map disease outbreaks, compare the effectiveness of treatments and improve understanding of conditions. Aggregated data from individual patient cases was crucial to containing the Ebola outbreak in West Africa.38 Private and public sector healthcare providers around the world are now using various forms of electronic medical records. These help individual patients by making it easier to personalise health services, but the public health benefits require these records to be interoperable. 

There is scope to launch collaborative projects to test the interoperability of data, standards and safeguards across the globe. The World Health Assembly’s consideration of a global strategy for digital health in 2020 presents an opportunity to launch such projects, which could initially be aimed at global health challenges such as Alzheimer’s and hypertension.39 

The slowing progress in bringing more people online points to the urgent need for new approaches to building digital infrastructure, a complex task that requires better coordination among many stakeholders: governments, international organisations, communications service providers, makers of hardware and software, providers of digital services and content, civil society and the various groups that oversee protocols and standards on which digital networks operate.

Improved digital cooperation on a data-driven approach to public health has the potential to lower costs, build new partnerships among hospitals, technology companies, insurance providers and research institutes and support the shift from treating diseases to improving wellness. Appropriate safeguards are needed to ensure the focus remains on improving health care outcomes. With testing, experience and necessary protective measures as well as guidelines for the responsible use of data, similar cooperation could emerge in many other fields related to the SDGs, from education to urban planning to agriculture. 

Data collaboration for climate change, agriculture and the environment 

The Platform for Big Data in Agriculture was launched in 2017 by the Colombia-based International Center for Tropical Agriculture after consultation with public, private and non-profit stakeholders. By providing ways to share data on agriculture, it seeks to transform research and innovation in food security, sustainability and climate change.40 

More broadly, cheaper sensors generating more data – and better AI algorithms to analyse it – can further improve our understanding of how complex environmental systems interact and the likely impacts of climate change.41 

Digital technologies can also be used to reduce waste. The methods of complex coordination that have lowered costs by enabling supply chains to touch every corner of the planet can also help to meet higher environmental standards and design devices with repair, reuse, upgrading and recycling in mind. For this, new forms of digital cooperation and data sharing would be needed among suppliers, customers and competitors


Many types of digital technologies and content – from data to apps, data visualisation tools to educational curricula – could accelerate achievement of the SDGs. When they are freely and openly available, with minimal restrictions on how they can be distributed, adapted and reused, we can think of them as “digital public goods”.42 In economics, a “public good” is something which anyone can use without charge and without preventing others from using it.43 Digital content and technologies lend themselves to being public goods in this respect.

Combinations of digital public goods can create “common rails” for innovation of inclusive digital products and services. The India Stack is an example of how a unified, multi-layered software platform with clear standards, provided by public entities, can give government agencies and entrepreneurs the technological building blocks to improve service delivery and develop new business models which promote economic inclusion.44 

There is currently no “go to” place for discovering, engaging with, building, and investing in digital public goods. Along the lines of the MOSIP model – and with the participation of civil society and other stakeholders – such a platform could create great value by enabling the sharing and adaptation of digital technologies and content across countries in a wider range of areas relevant to achieving the SDGs. 


The proportion of people online in the developing world expanded rapidly in the last decade – from 14.5% in 2008 to 45.3% in 2018 – but progress has recently slowed.45 Internet access in many parts of the world is still too slow and expensive to be effectively used.46 The cost of mobile data as a percent of income increased in nearly half the countries according to a recent study.47 Without affordable access, advances in digital technologies disproportionately benefit those already connected, contributing to greater inequality. 

The people being left behind are typically those who can least afford it. Growth in new internet connections is slowest in the lowest-income countries.48 Rural areas continue to lag, as companies prioritise improving access in more densely populated areas which will offer a better return on investment.49 

The slowing progress in bringing more people online points to the urgent need for new approaches to building digital infrastructure, a complex task that requires better coordination among many stakeholders: governments, international organisations, communications service providers, makers of hardware and software, providers of digital services and content, civil society and the various groups that oversee protocols and standards on which digital networks operate.50 As these actors cooperate, it also represents an important moment to re-emphasise and address the complex social, cultural and economic factors that continue to marginalise many groups. 

Some countries, such as Indonesia, have set targets that treat internet connectivity as a national priority.57 While finance alone will not achieve universal internet access, it can help if invested wisely: some countries are generating financing from fees on existing communication network providers to help expand systems to those who are currently uncovered, for example through Universal Service Funds.58

Advance market commitments deserve further consideration as a possible way to incentivise investment, as they have in other areas such as vaccine developments. They involve a commitment to pay for a future product or service once it exists; the commitment in this case could come from consortia of governments, international organisations or others interested in enabling specific uses in areas such as health or education.59

Many local groups are also working on small-scale community solutions: for example, a rural community of 6,000 people in Mankosi, South Africa, built a solar-powered “mesh network” in collaboration with a university.60 Such community projects are often not just about getting online but building skills and empowering locals to use technology for development and entrepreneurship.61

Digital cooperation should increase coordination among the public and private entities working in this space and help tailor approaches to economic, cultural and geographic contexts. Governments have an important role to play in creating a policy framework to enable private sector enterprise, innovation, and cooperative, bottom-up networks.


Even where getting online is possible and affordable, extra ‎efforts are needed to empower groups that are discriminated ‎against and excluded. For example, digital technologies are ‎often not easily accessible for elderly people or those with ‎disabilities;62 indigenous people have little digital content in their ‎native languages;63 and globally an estimated 12 percent more ‎men use the internet than women. 64

Even where getting online is ‎possible and affordable, extra ‎efforts are needed to empower ‎groups that are discriminated ‎against and excluded. ‎

Responses need to address deep and complex social and cultural factors, such as those contributing to the gender gap in access to and usage of mobile phones, smart phones and digital services – gaps which persist in many cases despite increases in women’s income and education levels.65 Social marketing could play a role in changing attitudes, as it has in many other areas with backing from donors, governments and civil society organisations.66 Initiatives to improve access for marginalised populations should start with consultation involving these groups in the design, deployment and evaluation of such efforts.

Efforts to improve digital inclusion would be greatly helped if there were a clear and agreed set of metrics to monitor it. Initial work – notably by the Organisation for Economic Co-operation and Development (OECD), the Group of Twenty (G20), ITU, and the Economist Intelligence Unit – needs to be broadened to reflect the wide variety of global contexts and, importantly, needs greater buy-in and participation from developing countries.67 The Panel urges international organisations, civil society and governments to develop action plans around reliable and consistent measures of digital inclusion with sex disaggregated data. Discussion about measurements and definitions would also focus attention on the issues underlying inclusion.


Many previous waves of technological change have shifted what skills are demanded in the labour market, making some jobs obsolete while creating new ones. But the current wave of change may be the most rapid and unpredictable in history. How to prepare people to earn a livelihood in the digital age – and how to protect those struggling to do so – is a critical question for digital cooperation for governments and other stakeholders who aim to reduce inequality and achieve the SDGs.

At this stage, there appears to be limited value in attempting to predict whether robots and artificial intelligence will create more jobs than they eliminate, although technology historically has been a net job creator.68 Many studies attempt to predict the impact on the jobs market but there is far from being a consensus.69 The only certainty is that workers have entered a period of vast and growing uncertainty – and that this necessitates new mechanisms of cooperation. 


Modern schools were developed in response to the industrial revolution, and they may ultimately need fundamental reform to be fit for the digital age – but it is currently difficult to see more than the broad contours of the changes that are likely to be needed.

Countries are still in early stages of learning how to use digital tools in education and how to prepare students for digital economies and societies. These will be ongoing challenges for governments and other stakeholders. Some countries are now exposing even very young children to science and robotics. Alongside such broader digital literacy efforts, it may be even more important to focus from an early age on developing children’s “soft skills”, such as social and emotional intelligence, creativity, collaboration and critical thinking. One widely referenced study concludes that occupations requiring such soft skills are less likely to be automated.70

Teaching about specific technologies should always be based on strong foundational knowledge in science and math, as this is less likely to become obsolete. At a degree level, science, technology, engineering and mathematics (STEM) curricula need to borrow from the humanities and social sciences, and vice versa: STEM students need to be encouraged to think about the ethical and social implications of their disciplines, while humanities and social science students need a basic understanding of data science.71 More informal approaches to learning may be needed to prepare students for working in cross-disciplinary teams, and where such informal approaches already exist in the developing world they should be fully appreciated for their value.

As the boundaries increasingly blur between ‘work’ and ‘learning’, the need to enable and incentivise lifelong learning was emphasised in many of the written contributions the Panel received.

Lifelong learning should be affordable, portable and accessible to all. Responsibility for lifelong learning should be shared between workers themselves, governments, education institutions, the informal sector and industry: digital cooperation mechanisms should bring these groups together for regular debates on what skills are required and how training can be delivered. Workers should have flexibility to explore how best to opt into or design their own approach to lifelong learning.

There are emerging examples of government efforts to use social security systems and public-private partnerships to incentivise and empower workers to learn new skills and plan for a changing labour market. Among those drawn to the Panel’s attention were efforts by the International Trade Union Confederation in Ghana and Rwanda,72 France’s Compte Personnel de Formation, Scotland’s Individual Training Account, Finland’s transformation of work and the labour market sub-group under its national AI programme, and Singapore’s Skills Framework for Information and Communication Technology (ICT).

However, reskilling cannot be the only answer to inequality in the labour market – especially as the workers most able to learn new skills will be those who start with the advantage of comparatively higher levels of education.73


New business models are fuelling the rise of an informal or “gig” economy, in which workers typically have flexibility but not job or income security.74 In industrialised countries, as more and more people work unpredictable hours as freelancers, independent contractors, agency workers or workers on internet platforms, there is an urgent need to rethink labour codes developed decades ago when factory jobs were the norm.75

Promising initiatives include Germany’s Crowdsourcing Code of Conduct, which sets out guidelines on fair payment, reasonable timing and data protection for internet platform workers, and employs an ombudsman to mediate disputes; and Belgium’s Titre-Services and France’s Chèque Emploi Service Universel, which offer tax incentives for people engaging casual workers to participate in a voucher scheme that enables the workers to qualify for formal labour rights. There are also examples of digital technologies enabling new ways for workers to engage in collective bargaining.76

While the gig economy tends to make work less formal in industrialised countries, in the developing world the majority of people have long worked in the informal sector.77 For these workers, gig economy arrangements may be more formal and transparent, and – with appropriate cooperation measures with technology firms – easier for governments to oversee.78 The challenge, as with industrialised countries, is to uphold labour rights while still allowing flexibility and innovation.

In all national contexts, protecting workers and promoting job creation in the digital age will require smart regulations and investments, and policies on taxation and social protection policies which support workers as they seek to transition to new opportunities.



Taxation, trade, consumer protection and competition are among the areas of economic policy that require new thinking in the digital age: they are the ‘guard rails’ of the digital economy. Increased cooperation could lead to effective national approaches and experience informing regional and global multilateral cooperation arrangements.

Taxation, trade, consumer protection and competition are among the areas of economic policy that require new thinking in the digital age: they are the ‘guard rails’ of the digital economy. Increased cooperation could lead to effective national approaches and experience informing regional and global multilateral cooperation arrangements.

Currently, however, there is a lack of regional and global standards in these areas, and multilateral cooperation is generally not working well. This may inflict far higher costs than is widely recognised. To take one relatively simple example, regional and global standards in areas such as interoperability of mobile money systems and best practices for digital ID would have considerable benefits. To discourage misuse, such standards and practices would also need to include clear accountability.

International trade rules need to be updated for the digital age. Technologies and trade have changed dramatically since 1998, for example, when the World Trade Organisation (WTO) last brokered an agreement on e-commerce.79 In January 2019, 76 WTO member states announced the initiation of plurilateral negotiations on trade-related aspects of e-commerce.80 Any agreement will need to address concerns of a diverse range of countries, including lower-income countries in which the e-commerce sector is less developed.81

Some argue that restrictions on data flows should be treated like any other trade barrier and generally minimised.83 However, views differ sharply, and decisions on national legislation are complicated by concerns about privacy and security – discussed in the next chapter. Countries that require companies to store and process data within their national borders argue that it promotes local innovation and investment in technology infrastructure and makes it easier to tax global corporations.84 Others argue against such approaches on the basis that they are protectionist or represent an effort to obtain access to the data.

There is growing recognition that taxation is an area where digital technology has moved faster than policy frameworks. In particular, technology firms may operate business models – such as multi-sided platforms or “freemium” models – which offer free services to some individual users and earn revenue from other users, merchants or advertisers.85 A company may provide services to millions of people in a country without establishing a legal entity or paying tax there. This has become a source of growing popular resentment.86

Where possible, new regulatory approaches should be tested on a small scale before being rolled out widely – through, for example, pilot zones, regulatory sandboxes or trial periods.

International digital cooperation could assist countries to develop appropriate tax policies. The G20 and OECD’s Base Erosion and Profit Shifting project is currently seeking consensus on issues such as how a global company’s tax receipts should be allocated to different jurisdictions based on its business activities.87 An agreement in this area could offer countries a source of revenue that they could, for example, use to invest in human capital or lower the tax burden on small businesses.

Some countries are now taking unilateral action. Countries such as Italy, France and the United Kingdom (UK) have announced the intent to impose taxes on digital sales rather than profits, at least on an interim basis.88 Other countries, such as Thailand, have amended tax rules relating to offshore digital services.89 The lack of cooperation and coordination among different regulators is creating a patchwork of different national rules and regulations which makes trade and e-commerce more difficult. Ensuring that such emerging tax policies do not have unintended consequences on small enterprises or poor populations deserves special attention.

An international perspective is also needed to tackle concerns about competition, which have grown as large firms have established leading positions in many digital services. This is due in part to network effects: the more users a platform already has, the more attractive it becomes for new users and advertisers.

Recent discussions have proposed three main approaches.90 First, a relatively laissez-faire approach that favours self-regulation or minimal regulation. Proponents argue that government regulation is often poorly conceived and counterproductive, harming innovation and economic dynamism. Critics counter that an overly hands-off approach has led to a concentration of market power in large firms and abuses of privacy that have sparked public and government concern.

A second approach calls for more active state intervention to set rules for digital companies. Experience in industrial policy shows that such an approach can either help or hinder depending on many factors, including regulators’ willingness and ability to engage varied stakeholders in a smart discourse to balance competing interests effectively.91

A third approach suggests regulating digital businesses as public utilities, analogous to railroads or electricity companies. The analogy is not an exact one, however, as physical infrastructure is easier to segment and harder to replicate than digital infrastructure and lends itself more easily to hosting competition among service providers. There is also dispute about how contestable are digital markets – that is, how vulnerable are the leading firms to new competitors. Moreover, traditional competition law operates far more slowly than changes in technology.


Finding the right approach in these areas will require not only different countries to work together, but also regulators in different government agencies. Models for how agencies can come together for peer-to-peer information sharing include the International Conference of Data Protection & Privacy Commissioners and the International Competition Network.92

Alongside existing models, new models of governance and cooperation may be needed. They will need to be multi-stakeholder, including the private sector, civil society and users. Their debates should be transparent and open to citizens, as modelled by Mexico’s National Institute for Transparency, Access to Information and Personal Data Protection.93

Where possible, new regulatory approaches should be tested on a small scale before being rolled out widely – through, for example, pilot zones, regulatory sandboxes or trial periods. We stress the overall need for a “systems” approach to cooperation and regulation that is multi-stakeholder, adaptive, agile and inclusive in Recommendation 5B.

However, regulators need to have sufficient resources and expertise to engage in such an approach – and the Panel’s consultations highlighted concern that many regulators and legislators have insufficient understanding of complex digital issues to develop and implement policies, engage with companies developing technologies and explain issues to the public.94 This increases the risk of regulations having unintended consequences.

There are several existing examples of initiatives to develop the capacity and understanding of public officials, from countries such as Israel,95 Singapore96 and the United Arab Emirates (UAE).97 But much more could be done, and the Panel’s Recommendation 2 envisages “digital help desks” which would broaden opportunities for officials and regulators to develop the skills needed for the smart governance that will be required to create inclusive and positive outcomes for all.