• Groundbreaking Research

    The Initiative on the Digital Economy is committed to breaking new ground in the study of the digital revolution and its profound impact on society, work, and the economy. We are developing a constant stream of research results on critical topics, organized around our four key pillars, or subject areas:

    • Productivity, Employment, and Inequality
    • New Digital Business Models
    • Big Data
    • Social Analytics

    Sinan Aral – Rebirth of the Social Sciences Through the Use of Digital Tools

    “We are on the precipice of a breakthrough in our understanding of human behavior,” says IDE co-founder Sinan Aral. With the vast amount of granular data now available about human interactions—from mobile phone location information to transaction data and social media activity logs—social scientists can now study hundreds of millions of people at a time. These large-scale “in the wild” experiments will allow researchers to answer nuanced questions about behaviors, from health choices to voting patterns to how people share information and choose products and services. “With this new knowledge, we can develop much more adaptive and personalized approaches and much more effective social policy,” says Aral. Massive, detailed data sets will also enable researchers to study how outcomes vary between different segments of the population and how they change over time.

    In his work, Aral has conducted numerous large-scale experiments –with MTV on voter mobilization, with the Praekelt Foundation in South Africa on HIV testing, and with Nike on diet and exercise patterns, among others.

    Aral also analyzes social influence and information flow in social networks and the impact of that information on different groups of people. “How can we take advantage of our new knowledge of social influence and how information cascades through populations?” asks Aral. If we can better understand how information travels, he says, we can begin to design “network interventions” that can promote positive behaviors or contain negative ones. From crisis response to disease management and from product development to viral marketing, “The applications are limitless.”

    Sinan’s primary research revolves around three projects:

    1. Understand how to create influence over changing behavior on a broad scale through peer to peer interaction and communication.
    2. Advance the science and process of large scale experimentation
    3. Using big data for positive and collective good, including de-escalation of violence, increased voting participation, public health, etc.

    Sandy Pentland – On Big Data, Social Analytics, and Personal Data

    In his research, Alex ‘Sandy’ Pentland, one of the world’s leading data scientists, asks, “How can we make a new, digital society that really works for humans?”

    Pentland’s work on big data examines everything from credit scores and financial indicators to mobility, community health, and data privacy. In one recent project, he has been studying reams of available financial data to determine whether there are better ways for banks to determine borrowers’ financial health and make better lending decisions. In the wake of the financial crisis, Pentland says, old methods of evaluating creditworthiness have made it hard for young, small businesses to get loans, a result that has contributed to the slow recovery. Which standards could be adjusted to help credit flow more freely while maintaining safe lending levels? A related topic of interest is perhaps the most-watched metric of national financial health: GDP. Pentland asks whether there might be a better number, one that reflects a more accurate and timely picture of not only a country’s financial health, but also social health.

    In addition to his study of financial metrics, Pentland is investigating what he calls “sustainable digital ecologies.” With the interconnections in our increasingly digital society multiplying, Pentland says, “the system is fragile in ways we don’t understand.” Through his work on the so-called New Deal on Data, Pentland urges individuals, governments, and corporations to share data for the public good, so that the benefits of sophisticated big data analysis—early identification of epidemics, for example, or emerging financial bubbles—can accrue to all, rather than a few. “How can we get people to work together so that they can better and more safely share information? How can we build a digital society that is win-win-win, rather than win-lose-lose?” he asks.

    A key feature of Pentland’s work is his emphasis on designing in-the-field studies in real time—studying data sharing among people in a particular Italian city, for example, or building an online trading platform and watching how traders interact with it. “I don’t just study things. I build things and use them to explore the future,” he says.

    Sandy’s focused research activities include the following projects:

    1. We need a `New Deal on Data' that democratizes data, giving citizens and consumers a say over its use and letting them participate in its value. Without these sorts of protections the value from big data will accrue only to those with the biggest computers and fattest digital pipes. Although it sounds idealistic, the major part of this vision is quite practical from a computer science point of view, and from a political viewpoint is already becoming real through the US Consumer Privacy Bill of Rights and the EU Data Protection acts. Moreover I am testing the commercial, legal, and human dimensions of these new protections in `Living Labs', which are communities that have agreed to try out this new way of living, in order to assess the effectiveness of this `New Deal on Data'.
    2. As more and more of the world becomes `datafied', giving us a `digital mirror' that reflects every nuance of the physical world, governance of interactions will become digital. Increasingly purchases, credit, logistics planning, negotiations, and almost all legal disputes will be settled in cyberspace without human intervention. If we are to avoid becoming mere cogs in a vast digital matrix governing the world we must design this new world to reflect human values and engage human strengths. We need to understand the `social physics' of human society and use it as the blueprint for our new digital environment.
    3. Perhaps the central driver of innovation is sharing of information, yet sharing of information can also endanger personal privacy and liberty. We need to change how we share information in order to alter the risk-reward ratio of sharing, making it safer to share. To do this I have built computer systems that provide very safe, auditable sharing, and deployed these systems in `Living Labs' to test their effect. When deployed in `Living Labs', which are communities that have agreed to live with these new systems, we can test them and assess their impact on the rate of community innovation.

    Erik Brynjolfsson – Deep Understanding, Improvement, and Measurement of Labor, Wage Inequality, and Economic Health

    In his recent book with IDE co-director Andrew McAfee, Erik Brynjolfsson explores The Second Machine Age—a core concept in his research. Brynjolfsson looks at the increasing digitization of our society and, while acknowledging and celebrating the dramatic impact on productivity and the many fascinating things we have been able to achieve with technology, studies its impact on labor and jobs. “We’re trying to understand how technology is changing the workforce,” he says. “Technology has rushed ahead, but there are implications in terms of social impact and the need for managers and educators to adjust, and those changes are slower.”

    Brynjolfsson is searching for ways in which the workforce and educational institutions can transform themselves to keep pace with—or at least remain within striking distance of—the dramatic advance of technology. “There is a real risk that we are not developing an economy that is sustainable in terms of job creation,” he says, pointing to the recent economic recovery, which saw companies return to profitability far faster than it has seen workers return to pre-recession employment levels.

    In one recent project, Brynjolfsson and his team have undertaken a systematic analysis of all the occupations in the United States, examined the capabilities required for each job, and determined which skills are being automated and which are in increasing demand. The findings could help shape educational policy and workforce training efforts. “We have an educational system designed for a 20th century economy,” says Brynjolfsson, with students trained to memorize and follow instructions and rules, an approach initially designed to prepare them for factory work that today is largely mechanized. Instead, Brynjolfsson argues for an emphasis on creative thinking and problem-solving skills that will enable students to go on to the sorts of jobs that can only be done by humans – and tend to offer higher financial and emotional rewards as well.

    Erik’s primary research revolves around three projects:

    1. Deeper understanding and documentation of labor and wage inequality, labor (and capital) demand.
    2. What changes in the economic system will improve the creation of jobs – by creation of a blue print, technology, and innovation index.
    3. Create a better method/process/beacon for accurately measuring the health of an economy – by developing large scale experiments (conjoint analysis) to measure what people really want and value.

    Andrew McAfee - Smart Strategies and Policies as Technology Races Ahead

    The 'computer revolution' has accelerated recently, and it's not just one revolution. After decades of frustratingly slow progress the field of artificial intelligence has in the past few years seen significant breakthroughs in natural language processing, image recognition, inference, and other fundamental areas. Autonomous vehicles and robots have made similarly large strides, sometimes literally. The work of innovation and knowledge creation has opened up immeasurably as hundreds of millions of people (soon to be billions) have come online with powerful devices. And Moore's Law keeps operating to make computing better and cheaper and big data keeps getting bigger. Any one of these advances would be a significant development for our businesses, economies, and societies; in combination they're bringing the biggest changes since the Industrial Revolution.

    IDE cofounder Andrew McAfee has begun to explore these changes in Race Against the Machine and The Second Machine Age, his two books with Erik Brynjolfsson, as well as in his other work. He believes that as digital technologies continue to advance quickly, it's important to follow them out of the lab and into the real world in order to observe their effects on performance, competition, and organization and to learn how to best combine the capabilities of people and machines. He takes a cue from science fiction author William Gibson's observation that "the future is already here; it's just not evenly distributed." In that case, a sound strategy is to combine firsthand observation and more systematic data in order to understand where the future is farthest along, and how leaders can best shape it.

    Andy's primary research falls into three areas:

    1. Quantifying the business impact of recent advances in artificial intelligence. Deep learning and other techniques have revitalized the field of machine learning and enabled rapid progress in many problem areas. What happens when these advances are deployed in an industrial setting? How much and how quickly do performance measures change? How do the roles of people involve?
    2. Researching and writing a business book about the second machine age. How should business leaders take advantage of the fact that technology is racing ahead? How should they rethink strategies, redesign processes, and change structures? How should they organize, innovate, and go to market differently? Andy and Erik will continue their joint investigation of how digital progress is changing the business world.
    3. Exploring the societal choices we'll face as we continue to digitize. Technology is racing ahead, which is the best economic news on the planet, but as it's doing so it's leaving a lot of people behind. Inequality is growing in the US (and around the world) and the average worker is losing ground. These realities have given rise to a huge range of policy suggestions covering everything from taxation to regulation to education. Which ones are best for the second machine age? Will we at some point need to fundamentally rethink the pillars of our current capitalist economic systems?