Accurately collected data means truth about people, economies and government. True and reliable data have a critical role in determining whether a nation takes the path towards freedom or oppression, prosperity or poverty. Martin C.W. Walker and Muneeb Sikander write that data can literally make the difference between life and death, especially in the context of developing countries.
Complex civilisations throughout history have recognised the importance of collecting and using accurate data. The Harappa civilisation that flourished over four thousand years ago in what is now modern-day Pakistan is noted for its urban planning, including sophisticated water supply and drainage systems. Some of the most notable artefacts are perfectly standardised weights that were used for measurement across the cities of the civilisation. As other civilisations grew in sophistication, the techniques used to gather data developed as well from the censuses of the Roman world to Doomsday Book of William the Conqueror.
The importance of data to development was recognised by the United Nations in 2014 with the publication of the report “A World that Counts: Mobilising the Data Revolution for Sustainable Development”. The report led, amongst other things, to the creation of the Global Partnership for Sustainable Development, which involves governments, NGOs and Big Tech firms such as Google, Facebook and Microsoft to improve collection and usage of data to support sustainable development. While many of the policies and programmes that have grown out of the report are positive for the developing world, a national data policy is about far more than sustainability. The right data, if collected accurately, means truth about people, economies and government. Those truths, in turn, have a critical role in determining whether a nation takes the path towards freedom or oppression, prosperity or poverty. Data can literally make the difference between life and death.
Designing a national data policy consists of a relatively small number of key elements. The first challenge in creating a data strategy to drive development is to be clear about its tangible objectives. Driving and measuring government policy in areas such as health and education is fundamental, but in the digital age so is creating a framework that supports the growth of digital businesses. Unknown to most Wall Street and City of London bankers, much of the modern financial world runs on software that is created in places such as Lebanon, Sri Lanka and Tunisia. Encouraging the digital economy also means getting the balance right in collaboration with Big Tech firms whose turnover can be higher than some countries’ GDP. Objectives also need to consider the value of data transparency in supporting better governance and privacy in defending civil liberties. All of this while not losing sight of the need to get the data basics right.
Making any significant decision without knowing how many people it impacts magnifies the risk of failure. It can also undermine the legitimacy of the state. Lebanon’s division of powers is based on the relative size its ethno-religious communities, as recorded in 1932. No census has been taken since. Even the United States saw major battles regarding the Trump administration’s desire to ask residents if they were citizens in the 2020 census. The concept of a census is perhaps a relic of a pre-digital age. A more effective means of measuring the population is a comprehensive national identity system with unique identifiers for all citizens and residents. A way of uniquely identifying all persons and corporations is the backbone of any national data policy.
From the ability to uniquely identify people and companies grows the ability to create joined up systems for tax, health, education, voting and law enforcement. The ability to easily cross-reference different data sets allows far more timely snapshots of the population as a whole and can help tackle major issues in developing countries. For instance, a fundamental problem for most developing nations is collecting enough tax to fund public expenditures. There are many types of problems that include tax evasion, money laundering and fraud, but all fundamentally exist in the gaps between data sets. If one can join the data sets together, it is possible to apply new techniques such as machine learning to identify problems previously unknown in order to develop both new and novel ways to fix them.
Transparency and good governance
While the detail of much information gathered by governments is highly sensitive and needs to be restricted on privacy grounds, there is a general need to make data available to companies, academics and individual citizens. A common model of encouraging this is some form of freedom of information legislation that gives citizens at least the right to ask public bodies for access to any information that is not deemed secret or confidential. Creating an act to give citizens this right is one thing but being able to implement it in practice is quite another. There are two factors that can easily make a Freedom of Information Act a piece of worthless window dressing. One is the cost of enforcement for public servants to comply with the law. There is usually a cost ceiling on information requests, so it is vital that systems installed in the public sector are designed to allow the efficient extraction of information. It is therefore essential to have a fully empowered independent body existing outside the court system (which can be prohibitively slow and expensive in many developing countries) to make public servants provide information.
Big data versus good data
It is easy to get swept up in the hype over big data without realising that seldom does big data consist of the most valuable data. Understanding how direct and indirect variables drive outcomes requires being able to integrate different types of data. The tools built by firms such as Google, Amazon and Facebook, for instance, help to manage vast amounts of data, processing all those likes, shares and videos, because each individual data item is very low in value compared to, say, an individual’s date of birth or annual income. Big data is no substitute for key items of good data, but it can be a valuable supplement and check on the validity of official data. An official census may exclude unwanted minorities or refugees, but they are unlikely to be hidden in mobile phone statistics.
Is big tech our friend or foe?
Big tech firms are generous providers of support to many data-related projects in the developing world, but they also have an eye to generating large profits similar to large multinationals who have both helped and exploited such countries in the past. In spite of the power and apparent good wishes that big tech might exhibit, the governments of developing nations should aim to legally control the data generated and distributed within their borders, similar to how such legal frameworks are established in any developed nation. Respecting developing nation laws applying to privacy, intellectual property and advertising (amongst many others) should be enforced as the price of doing business for big tech. Developing nations must confront the reality that there is no such thing as a free lunch in economics and hence must weigh up the benefits that big tech offers vis-à-vis data ownership and privacy laws.
To fully realise the benefits of a data strategy we must first get the basics right. This means being more effective in how we collect, curate, store, manage and delete data. Furthermore, a mechanism that allows data to be shared across systems is critical. Finally, in order to change organisations and society, data requires effective governance, management and stewardship.
- The post expresses the views of its authors, and do not necessarily represent those of LSE Business Review or the London School of Economics.
- Featured image by insspirito, under a Pixabay licence
- When you leave a comment, you’re agreeing to our Comment Policy