The Data Economy

Implications from Singapore

Sree Kumar, Warren Chik, See-Kiong Ng, Sin Gee Teo

• Publisher: Taylor & Francis
• Year: 2018
• Print ISBN: 9781138359574
• eBook ISBN: 9780429782633
• Edition: 1
• Topic: Social Sciences
• Subtopic: Ethnic Studies
• Index: Social Sciences

"The data economy" is a term used by many, but properly understood by few. Even more so the concept of "big data". Both terms embody the notion of a digital world in which many transactions and data flows animate a virtual space. This is the unseen world in which technology has become the master, with the hand of the human less visible. In fact, however, it is human interaction in and around technology that makes data so pervasive and important – the ability of the human mind to extract, manipulate and shape data that gives meaning to it.

This book outlines the findings and conclusions of a multidisciplinary team of data scientists, lawyers, and economists tasked with studying both the possibilities of exploiting the rich data sets made available from many human–technology interactions and the practical and legal limitations of trying to do so. It revolves around a core case study of Singapore's public transport system, using data from both the private company operating the contactless payment system (EZ-Link) and the government agency responsible for public transport infrastructure (Land Transport Authority). In analysing both the possibilities and the limitations of these data sets, the authors propose policy recommendations in terms of both the uses of large data sets and the legislation necessary to enable these uses while protecting the privacy of users.

1Introduction

This book is a product of its time. The “data economy” has now become a common term, used by many and understood by few. There is a sense of mystery that shrouds the idea of a data economy, more so the concept of “big data.” Both terms embody the notion of a digital world in which many transactions and data flows animate virtual space. This is the unseen world in which technology has become the master with the hand of the human less visible. However, this is to miss the wood for the trees. The human interaction in and around technology is what makes data so pervasive and important. It is the ability of the human mind to extract, manipulate, and shape data that gives meaning to it. This transformation of data into meaningful information for decision-making is central to the idea of a “data economy,” and it involves the use of large and complex sets of data (i.e. big data).

This study is unusual in that it brings together a multidisciplinary team of data scientists, lawyers, and economists to construct a possible structure of a “data economy” and its evolution. From looking at how the use of data has expanded and how different entities such as governments, businesses, and individuals continue to generate and use data through various means, the study constructs the critical elements that inform the new economy. It draws on data from the transportation network¹ in Singapore to model travel patterns, number of footfalls at the different critical nodes by time of day, and segmentation of the data by parameters such as age group, gender, and tourist status. While these empirical findings are salient to the study, equally important has been the due process of obtaining data from different sources. The acquisition of data from public (Land Transport Authority [LTA]) and private (EZ-Link) sources has had its own rhythm, since they are governed by different legislative constructs. This process has also highlighted how the integration of different data sets, their curation, and their standardisation for analytics have unique complexities. These have been challenging in both their technical and legal demands. The legal requirements for confidentiality and the maintenance of privacy have, in their own ways, been strong filters through which the data sets have been obtained, anonymised, and structured for analytics. The results of these stringent demands have nevertheless been to support a robust analytics framework through which to visualise the complex movement of people through the transportation network in Singapore.

The study, while focusing on Singapore, has also drawn lessons from other jurisdictions. The use of data analytics for decision-making in urban activities is not new, but the growing cities of Asia have been at the forefront of using data in innovative ways. Seoul, Taipei, Hong Kong, Shanghai, and Tokyo are all well versed in the use of big data for urban management – even more so since these cities have embedded and embraced cashless payment systems for a variety of public and private transactions. The pool of data that they have collected is immense and some, such as Seoul, have used it in unusual ways for transport planning. Likewise, the Hong Kong transportation system (MTR Corp.) has tried to commercialise its customer database, resulting in a contested space between private and public interests. Thus, there are interesting features of data management, processing, and controls that are being explored in Asia and its metropolitan areas. There is now a rich vein of information regarding how data is being obtained and managed in Asia’s urban conurbations, and this study has tapped into this knowledge base.

One of the aims of this study has been to ensure that the latest in “big data” and data analytics is being tracked. Team members have therefore attended workshops and conferences on the latest developments in these areas in the United States and Israel. Some of the ideas from these meetings have found a voice in this study and have been integrated into the analytics framework used on the data sets.

The findings in this study reflect the different elements of how data can, and eventually will, become a part of economic structure. There are already signs of change in traditional economic activities as data become more important features of business and daily life. Online shopping, Internet banking, applications in transport such as Uber, or short-term lets through Airbnb are all acting in concert to change the ways in which traditional economic activity is structured. These are, in the parlance of the technology pundits, “disruptors.” While these new technology-based businesses are disrupting the marketplace, they are also redefining the deeper concepts of what a market will look like. Traditional economic theory posits the important role of the factors of production and how they are governed by rules and regulations to ensure an orderly marketplace. Land, labour, and capital were seen to be the critical factors of production in a traditional economy before being observed to have synergies that could give rise to new skills and improved productivity. As capital became better utilised through technology and innovation, the service economy emerged to become a more prevalent feature of modern society. Technology thus has facilitated the movement of large numbers from the land and from rural factories to seek livelihoods in the cities.

It is thus in the urban setting that the rules of engagement in the new marketplace are being contested and redrawn. Increasingly, data is a resource that is being used to fashion new demand and supply contours in the economy. Big data with appropriate analytics, for example, allows businesses to segment their customers, define their product lines with more precision in quality and pricing, and focus their sales to a better-defined customer base. In turn, customers have now become more discerning in their tastes as the Internet and applications on their mobile phones allow them to compare quality, design, and prices from different suppliers without leaving the home, the office, or their trains and buses. This narrowing in the asymmetry of information has its own challenges, including that of managing expectations of consumers and addressing the risks of misinterpreting vital information. This change in the structure of information can be easily ascertained in the case of medical products and medications that a supposedly “informed” consumer purchases for self-medication without proper medical supervision. Thus, the availability of data in various forms – digital, textual, visual, audio – raises several prospects for their capture, processing, manipulation, and eventual use in decision-making. The role of data analytics has, as a result, become a critical feature in the use of big data in business, policy-making, and personal decision-making.

These emerging trends portray the arrival of data as a constituent of the economic landscape. It is now a raw input into algorithms (the analytics part of the process), which then massages and manipulates that input to create value-added information. The information itself may then have to be consolidated into different frameworks so that meaningful insights can be derived from that collage. There may be significant value embedded in those insights that have come about as a result of several factors – data inputs, data processing, information extraction, framework definition and creation, and finally a human interpretation of the patterns that have emerged (i.e. the insight into a particular solution to a problem or into an understanding of the problem itself). This can best be illustrated through an example such as footfalls in a shopping mall. If data on the movement of people in a shopping mall throughout the day can be captured through an electronic gateway or a security camera, this data can be curated, verified, and structured for data analytics in order to extract important parameters such as number of footfalls per hour, gender, age, and the like. If the framework of the subsequent analysis is to identify the peak periods and the profile of the consumers, then this information can be culled through the analytics to become a part of marketing inputs for the mall. There is an inherent value to this information, in that the mall owners can design an appropriate strategy to attract more customers. If there is an increase in the number of customers as a result of this exercise, then this change in customer numbers and the attendant change in revenues captured by the retailers in the mall are reflections of the imputed value of the data, their analytics, and the interpretation of the underlying patterns.

This simple example also raises a further possibility in economic expansion. The process of collecting, verifying, and curating data, as well as running analytics, is a specialised function that requires unique skills. The idea of a data scientist is now a reality as industry and services have realised the value of data and the need to extract important insights from it. So data scientists have become sought-after specialists. The advent of cloud computing, faster processing capabilities, and machine learning have coalesced to provide data scientists with a new armoury with which to address the requirements of the marketplace. The growth of a new sector – data analytics and management – is a consequence of the widespread availability and use of data by business and government. In a similar fashion, there is also the corresponding growth in new service sectors such as cybersecurity and data protection.

The overall effect of data becoming a resource like any other and that can be used as an input for value creation is to nurture new economic sectors while allowing the current economic space to become more efficient and, often, narrowing its contours by removing obsolete activities. This Schumpeterian world of creative destruction has become more rapid as technology has diffused into many more areas of work, leisure, and production. As a result of these changes, the nature of work itself is undergoing a structural shift. More is being achieved through less physical effort, while intellectual endeavour has become more entrenched – “intellectual” in the sense that the mind has to become flexible and attuned to several tasks requiring analytical capabilities. There is thus a reduced need for physical labour and a more pronounced requirement for knowledge-intensive workers willing to participate in a digital world. This restructuring of the workforce carries with it several social responsibilities, including a need to narrow the digital divide, facilitate seamless movement across sectors, and enhance the opportunities for skills improvement. The role of public policy becomes all the more salient in this amorphous and rapidly shifting environment.

It can thus be observed that the data economy has a wide canvas of interaction within its ambit. It affects business, government, the labour market, and the consumer in a variety of ways. The interaction between the public and private spheres of activity brings to the fore the need for protecting the public interest as much as that of the individual. There is a legislative imperative to ensure that the rules and regulations that define the marketplace are well designed to prevent abuses and breaches of privacy. The nature of the data economy is such that there can be a high degree of privacy loss, abuse of the due processes of governance, and the potential for a loss in trust in the institutions that govern the market. Thus, the legal contours that define the marketplace must keep pace with changes in technology and the manner in which data is being acquired for analytics and its subsequent use. The value chain of the information creation process must be policed effectively to prevent unlawful use and to maintain the privacy and integrity of the data that has been collected. These are onerous demands in a technology-intensive environment in which skills and capabilities for data management are in short supply. Thus, the legal perimeters of the market have to become much better defined, transparent, and neutral to allow the marketplace to function effectively.

What is becoming more visible in the new economy is the changing share of production and usage of data by the various agents – governments, private entities, and individuals. It is likely that the share of data produced and used by private agents and individuals will increase and overtake that which is contributed by governments. This is to be expected as non-government entities experiment, commercialise, and widen their reach to capture a much larger audience. However, this expansion will depend on how technology deepens and becomes cost-effective in collecting data, analysing it, and disseminating the resulting information. Innovation in technology and its use will therefore be a catalyst for this process to unfold. This also implies that a widening of the current skills base will be necessary so that technology is used most effectively. The calls for specialised skills within data science and its cluster of activities will increase significantly even as machine learning and artificial intelligence (AI) become more prevalent. This proliferation of data use and the attendant creation of new economic activities will call for better data protection, increasing demands for privacy, and a fairer marketplace. These are activities that have a significant bearing on public space and will require a stronger hand of governments. As a result, the role of governments in legislating, regulating, and overseeing the market will increase in importance. The failure of self-regulation, as has been seen in the past, is an indicator of how intensive this intervention will need to be in order to balance public and private interests. A new concord of rules and regulations that govern the data economy may be the outcome of this need for greater policing of the new virtual world.

This book explores these possibilities and defines some of these changes in the light of the empirical evidence obtained from the Singapore transportation data. More so, it also looks at the legal contours that define these activities through lessons learnt from other jurisdictions. Taken together, the results of this study posit the evolution of the data economy in the context of a city economy and the challenges it poses for decision-makers.

Note

1Data has been provided by the Land Transport Authority (LTA) and its commercial subsidiary, EZ-Link, for this study.

2An outline of the data economy

Data has now become a fixed feature of most economic activities. It is being captured through various means – digital, textual, visual, and, in some cases, audio technologies. The amount of data collected is vast compared to the amount that is actually analysed and used for decision-making. This is not surprising since the tendency to collect is often all-encompassing because of the uncertainty associated with how useful data may be in the future. As a result, there is much more data available in different forms than is actually analysed and acted upon. Nevertheless, this large pool of data has now become a form of raw “input” for sophisticated analysis from which deep insights can be gained. There are now comparisons with how oil became a feedstock for many of the world’s industries.¹

Just as oil is the base for many industrial activities, data has now become the input for many services and industries that are interconnected through digital pathways. But the comparison with oil or other such similar input runs into some obstacles when it is observed that valuing and establishing ownership of data can sometimes be more difficult than in the case of physical products. This is partly due to the nature of data that is not readily countable, the influence of network effects, and the presence of information asymmetries. However, if data is to be seen as an integral part of the new economy, then it must also exhibit, or be a part of, some of the features of an economic system. This chapter will look at several elements that constitute a structure that can be termed a “data economy.”

The elements of a data economy

A traditional economy is defined as having the factors of production – land, labour, and capital – being used in a formulaic manner so as to create outputs from specific input materials. A market for these outputs (and for inputs as well) is then created for beneficial exchange through transactions that can be barter trade or, in a more sophisticated case, through a monetary exchange of some form. As the number of producers changes, such as through mergers or acquisitions, there may arise oligopolies or monopolies that arise to control the quantity and price of products. The same can be true in the provision of services that use labour and capital in some unique combination. In both instances, government intervention becomes necessary to prevent abuse of market power and to protect the public interest, which in this case will also include the many other smaller producers who are also consumers. However, governments also intervene to ensure that the public interest is protected for other reasons as well, such as safety, health, and orderly market behaviour. This rudimentary example highlights the different participants in an economy – producers, consumers, and governments – and also the ways in which the hand of government may be an important feature of managing the marketplace. But the fundamental assumption in this example is also that the raw materials for production are also available through some market mechanism that exists within or outside the immediate ...

Table of contents

• Cover
• Half Title
• Series Information
• Title Page
• Copyright Page
• Contents
• List of illustrations
• Foreword
• Acknowledgements
• 1 Introduction
• 2 An outline of the data economy
• 3 The search for data
• 4 The analysis of data
• 5 The legal contours of the data economy
• 6 Redefining the data economy
• 7 Implications for the future
• Selected bibliography
• Index