“Missed connections” is the personals ads category for people whose encounters are too fleeting to form any union – a lost-and-found for relationships.  I gave that title to my paper on the conversation between the United States and for Europe on data, privacy, and surveillance because I thought it provides an apt metaphor for the hopes and frustrations on both sides of that conversation.

The United States and Europe are linked by common values and overlapping heritage, an enduring security alliance, and the world’s largest trading relationship.  Europe has become the largest crossroad of the Internet and the transatlantic backbone is the global Internet’s highest capacity route.

[I]

But differences in approaches to the regulation of the privacy of personal information threaten to disrupt the vast flow of information between Europe and the U.S.  These differences have been exacerbated by the Edward Snowden disclosures, especially stories about the PRISM program and eavesdropping on Chancellor Angela Merkel’s cell phone.  The reaction has been profound enough to give momentum to calls for suspension of the “Safe Harbor” agreement that facilitates transfers of data between the U.S. Europe; and Chancellor Merkel, the European Parliament, and other EU leaders who have called for some form of European Internet that would keep data on European citizens inside EU borders.  So it can seem like the U.S. and EU are gazing at each other from trains headed in opposite directions.

My paper went to press before last week’s European Court of Justice ruling that Google must block search results showing that a Spanish citizen had property attached for debt several years ago.  What is most startling about the decision is this information was accurate and had been published in a Spanish newspaper by government mandate but – for these reasons – the newspaper was not obligated to remove the information from its website; nevertheless, Google could be required to remove links to that website from search results in Spain. That is quite different from the way the right to privacy has been applied in America.  The decision’s discussion of search as “profiling” bears out what the paper says about European attitudes toward Google and U.S. Internet companies.  So the decision heightens the differences between the U.S. and Europe.

Nonetheless, it does not have to be so desperate.  In my paper, I look at the issues that have divided the United States and Europe when it comes to data and the things they have in common, the issues currently in play, and some ways the United States can help to steer the conversation in the right direction.

[I] "Europe Emerges as Global Internet Hub," Telegeography, September 18, 2013.

       

The United States exports digital goods worth hundreds of billions of dollars across the Atlantic each year.  And both Silicon Valley and Hollywood do big business with Europe every year.  Differences in approaches to privacy have always made this relationship unsteady but the Snowden disclosures greatly complicated the prospects of a Transatlantic Trade and Investment Partnership.  In this paper Cameron Kerry examines that politics of transatlantic trade and the critical role that U.S. privacy policy plays in these conversations.

Kerry relies on his experience as the U.S.’s chief international negotiator for privacy and data regulation to provide an overview of key proposals related to privacy and data in Europe.  He addresses the possible development of a European Internet and the current regulatory regime known as Safe Harbor. Kerry argues that America and Europe have different approaches to protecting privacy both which have strengths and weaknesses.

To promote transatlantic trade the United states should:

• Not be defensive about its protection of privacy
• Provide clear information to the worldwide community about American law enforcement surveillance
• Strengthen its own privacy protection
• Focus on the importance of trade to the American and European economies

Downloads

• Download the paper

       

       

In the past decade, a remarkable shift has occurred in the development landscape. Specifically, acknowledgment of the central role of the private sector in contributing to, even driving, economic growth and global development has grown rapidly. The data on financial flows are dramatic, indicating reversal of the relative roles of official development assistance and private financial flows. This shift is also reflected in the way development is framed and discussed, never more starkly than in the Addis Abba Action Agenda and the new set of Sustainable Development Goals (SDGs). The Millennium Development Goals (MDGs), which the SDGs follow, focused on official development assistance. In contrast, while the new set of global goals does not ignore the role of official development assistance, they reorient attention to the role of the business sector (and mobilizing host country resources).

The U.S. Agency for International Development (USAID) has been in the vanguard of donors in recognizing the important role of the private sector to development, most notably via the agency’s launch in 2001 of a program targeted on public-private partnerships (PPPs) and the estimated 1,600 USAID PPPs initiated since then. This paper provides a quantitative and qualitative presentation of USAID’s public-private partnerships and business sector participation in those PPPs. The analysis offered here is based on USAID’s PPP data set covering 2001-2014 and interviews with executives of 17 U.S. corporations that have engaged in PPPs with USAID.

The genesis of this paper is the considerable discussion by USAID and the international development community about USAID’s PPPs, but the dearth of information on what these partnerships entail. USAID’s 2014 release (updated in 2015) of a data set describing nearly 1,500 USAID PPPs since 2001 offers an opportunity to analyze the nature of those PPPs.

On a conceptual level, public-private partnerships are a win-win, even a win-win-win, as they often involve three types of organizations: a public agency, a for-profit business, and a nonprofit entity. PPPs use public resources to leverage private resources and expertise to advance a public purpose. In turn, non-public sectors—both businesses and nongovernmental organizations (NGOs)—use their funds and expertise to leverage government resources, clout, and experience to advance their own objectives, consistent with a PPP’s overall public purpose. The data from the USAID data set confirm this conceptual mutual reinforcement of public and private goals.

The goal is to utilize USAID’s recently released data set to draw conclusions on the nature of PPPs, the level of business sector engagement, and, utilizing interviews, to describe corporate perspectives on partnership with USAID.

The arguments regarding “why” PPPs are an important instrument of development are well established. This paper presents data on the “what”: what kinds of PPPs have been implemented and in what countries, sectors, and income contexts. There are other research and publications on the “how” of partnership construction and implementation. What remains missing are hard data and analysis, beyond the anecdotal, as to whether PPPs make a difference—in short, is the trouble of forming these sometimes complex alliances worth the impact that results from them?

The goal of this paper is not to provide commentary on impact since those data are not currently available on a broad scale. Similarly, this paper does not recommend replicable models or case studies (which can be found elsewhere), though these are important and can help new entrants to join and grow the field. Rather, the goal is to utilize USAID’s recently released data set to draw conclusions on the nature of PPPs, the level of business sector engagement, and, utilizing interviews, to describe corporate perspectives on partnership with USAID.

The decision to target this research on business sector partners’ engagement in PPPs—rather than on the civil society, foundation, or public partners—is based on several factors. First, USAID’s references to its PPPs tend to focus on the business sector partners, sometimes to the exclusion of other types of partners; we want to understand the role of the partners that USAID identifies as so important to PPP composition. Second, in recent years much has been written and discussed about corporate shared value, and we want to assess the extent to which shared value plays a role in USAID’s PPPs in practice.

The paper is divided into five sections. Section I is a consolidation of the principal data and findings of the research. Section II provides an in-depth “data picture” of USAID PPPs drawn from quantitative analysis of the USAID PPP data set and is primarily descriptive of PPPs to date. Section III moves beyond description and provides analysis of PPPs and business sector alignment. It contains the results of coding certain relevant fields in the data set to mine for information on the presence of business partners, commercial interests (i.e., shared value), and business sector partner expertise in PPPs. Section IV summarizes findings from a series of interviews of corporate executives on partnering with USAID. Section V presents recommendations for USAID’s partnership-making.

Downloads

• WP94PPPReport2016Web

      

Publish What You Fund’s 2016 Aid Transparency Index is out. And as a result, today we can assess whether major donors met the commitments they made five years ago at Busan to make aid transparent by the end of 2015. The index is also a window into the state of foreign aid transparency and how the U.S.—the world’s largest bilateral donor—stacks up.

The global picture

On the positive side, the index found that ten donors of varied types and sizes, accounting for 25 percent of total aid, have met the commitment to aid transparency. And more than half of the 46 organizations included in the 2016 index now publish data to the International Aid Transparency Initiative (IATI) registry at least quarterly.

At the same time, the index’s assessments show more than half of the organizations still fall into the lowest three categories, scoring below 60 percent in terms of the transparency of their information.

The U.S. picture

Continuing its leadership on transparency, the Millennium Challenge Corporation comes in second overall in the index, meeting its Busan commitment and once again demonstrating that the institutional commitment to publishing and using its data continues.

Otherwise, at first glance, U.S. progress seems disappointing. The five other U.S. donors included in the 2016 index are all in the “fair” category. Seen through a five-year lens, however, these same five U.S. donors were either in the “poor” or “very poor” categories in the 2011 index. So, all agencies have moved up, and three of them—U.S. Agency for International Development (USAID), Department of the Treasury, and the U.S. President's Emergency Plan for AIDS Relief—are on the cusp of “good.”

In the two biggest U.S. agencies that administer foreign assistance, USAID and the State Department, the commitment is being institutionalized and implemented through more systematic efforts to revamp their outdated information systems. Both have reviewed the gaps in their data reporting systems and developed a path forward. USAID’s Cost Management Plan identifies specific steps to be taken and is well under way. The State Department Foreign Assistance Data Review (FADR) involves further reviews that need to be executed promptly in order to lead to action. Both are signs of a heightened commitment to data transparency and both require continued agency leadership and staff implementation.

The Department of Defense, which slid backwards in the last three assessments (and began at the "very poor" category in 2011), has for the first time moved into the "fair" category.  It is still the lowest performing U.S. agency in the index, but it is now publishing 12 new IATI fields. It is moving in the right direction, but significant work remains to be done.

The third U.S. National Action Plan (NAP) announced last fall—the strongest issued by the U.S. to date—calls for improvements to quality and comprehensiveness of U.S. data and commits the U.S. to doing more to raise awareness, accessibility, and demand for foreign assistance data. This gives all U.S. agencies the imperative to do much more to make their aid information transparent and usable.  

Going forward—what should the U.S. being focusing on?

The overall challenge has been laid out in the third NAP:

• Almost all of the U.S. agencies need to improve the breadth and depth of the information they are publishing to meet IATI standards. Far too often, basic information—such as titles—are either not published or are not useful.
• The Millennium Challenge Corporation should continue its leadership role, especially on data use. All agencies should be promoting the use of data among their own staff and by external stakeholders, especially at country level. Feedback will go a long way toward helping them improve the quality of the data they are publishing and thereby help them meet the IATI standards.
• USAID must finish the work on its Cost Management Plan, including putting IATI in the planned Development Information Solution. Additionally, more progress needs to be made on the follow-up to the Aid Transparency Country Pilot Assessment to meet the needs of partners.   
• The State Department needs to follow through on including IATI in the new integrated solution mapped out in its data review.

The leadership of all foreign affairs agencies needs to work harder to make the business case for compiling, publishing, and using data on foreign aid programs. Open data, particularly when it is comparable, timely, accessible, and comprehensive, is an extremely valuable management asset.  Agency leadership should be its champion. So far, we have not seen enough.

U.S. progress on aid transparency was slow to start. It is still not where it needs to be. But with a modest but concerted push, three additional agencies will be in the “good” category and that is a story we can start to be proud of.   

We look forward to continued progress and to the day when all U.S. foreign aid meets transparency standards—a day I believe will be an important one for the cause of greater development, better governance, democratic participation, and reduced poverty worldwide.

       

       

       

       

       

       

       

There is a lot of hope around the potential of Big Data—massive volumes of data (such as cell phone GPS signals, social media posts, online digital pictures and videos, and transaction records of online purchases) that are large and difficult to process with traditional database and software techniques—to help achieve the sustainable development goals. The United Nations even calls for using the ongoing Data Revolution –the explosion in quantity and diversity of Big Data—to make more and better data usable to inform development analysis, monitoring and policymaking: In fact, the United Nations believes that that “Data are the lifeblood of decision-making and the raw material for accountability. Without high-quality data providing the right information on the right things at the right time; designing, monitoring and evaluating effective policies becomes almost impossible.” The U.N. even held a “Data Innovation for Policy Makers” conference in Jakarta, Indonesia in November 2014 to promote use of Big Data in solving development challenges.

Big Data has already played a role in development: Early uses of it include the detection of influenza epidemics using search engine query data or the estimation of a country’s GDP by using satellite data on night lights. Work is also under way by the World Bank to use Big Data for transport planning in Brazil.

During the Data for Development session at the recent NetMob conference at MIT, we presented a paper in which we jump on the Big Data bandwagon. In the paper, we use mobile phone data to assess how the opening of a new toll highway in Dakar, Senegal is changing how people commute to work (human mobility) in this metropolitan area. The new toll road is one of the largest investments by the government of Senegal and expectations for its developmental impact are high. In particular, the new infrastructure is expected to increase the flow of goods and people into and out of Dakar, spur urban and rural development outside congested areas, and boost land valuation outside Dakar. Our study is a first step in helping policymakers and other stakeholders benchmark the impact of the toll road against many of these objectives.

Assessing how the impact of the new toll highway differs by area and how it changes over time can help policymakers benchmark the performance of their investment and better plan the development of urban areas.

The Dakar Diamniadio Toll Highway

The Dakar Diamniadio Toll Highway (in red in Figure 1), inaugurated on August 1, 2013 is the first section (32 km or 20 miles) of a broader project to connect the capital, Dakar, through a double three-lane highway to a new airport (Aeroport International Blaise Diagne, AIBD) and a special economic zone, the Dakar Integrated Special Economic Zone (DISEZ) and the rest of the country.

Note: The numbers indicate the incidence of increased inter cell mobility and were used to calculate the percentage increase in mobility.

The cost of this large project is estimated to be about $696 million (FCFA 380.2 billion or 22.7 percent of 2014 fiscal revenues, excluding grants) with the government of Senegal having already disbursed $353 million. The project is one of the first toll roads in sub-Saharan Africa (excluding South Africa) structured as a public-private partnership (PPP) and includes multilateral partners such as the World Bank, the French Development Agency, and the African Development Bank.

In our study, we ask whether the new toll road led to an increase in human mobility and, if so, whether particular geographical areas experienced higher or lower mobility relative to others following its opening.

Did the Highway Increase Human Mobility?

Using mobile phone usage data (Big Data), we use statistical analysis in our paper to approximate where people live and where they work. We then estimate how the reduction in travel time following the opening of the toll road changes the way they commute to work.

As illustrated in the map of Figure 1, we find some interesting trends:

• Human mobility in the metropolitan Dakar area increased on average by 1.34 percent after the opening of the Dakar Diamniadio Toll Highway. However, this increase masks important disparities across the different sub-areas of the Dakar metropolitan areas. Areas in blue in Figure 1 are those for which mobility increased after the opening of the new road toll while those in red experienced decreased mobility.
• In particular, the Parcelles Assainies suburban area benefited the most from the toll road with an increase in mobility of 26 percent. The Centre Ville (downtown) area experienced a decrease in mobility of about 20 percent.

These trends are important and would have been difficult to discover without Big Data. Now, though, researchers need to parse through the various reasons these trends might have occurred. For instance, the Parcelles Assainies area may have benefited the most because of its closer location to the toll road whereas the feeder roads in the downtown area may not have been able to absorb the increase in traffic from the toll road. Or people may have moved from the downtown area to less expensive areas in the suburbs now that the new toll road makes commuting faster.

The Success of Big Data

From these preliminary results (our study is work in progress, and we will be improving its methodology), we are encouraged by the fact that our method and use of Big Data has three areas of application for a project such as this:

Benchmarking: Our method can be used to track how the impact of the Dakar Diamniadio Toll Highway changes over time and for different areas of the Dakar metropolitan areas. This process could be used to study other highways in the future and inform highway development overall.

Zooming in: Our analysis is a first step towards a more granular study of the different geographic areas within the Dakar suburban metropolitan area, and perhaps inspire similar studies around the continent. In particular, it would be useful to study the socio-economic context within each area to better appreciate the impact of new infrastructure on people’s lives. For instance, in order to move from estimates of human mobility (traffic) to measures of “accessibility,” it will be useful to complement the current analysis with an analysis of land use, a study of job accessibility, and other labor markets information for specific areas. Regarding accessibility, questions of interest include: Who lives in the areas most/least affected? What kind of jobs do they have access to? What type of infrastructure do they have access to? What is their income level? Answers to these questions can be obtained using satellite information for land prices, survey data (including through mobile phones) and data available from the authorities. Regarding urban planning, questions include: Is the toll diverting the traffic to other areas? What happens in those areas? Do they have the appropriate infrastructure to absorb the increase in traffic?

Zooming out: So far, our analysis is focused on the Dakar metropolitan area, and it would be useful to assess the impact of new infrastructure on mobility between the rest of the country and Dakar. For instance, the analysis can help assess whether the benefits of the toll road spill over to the rest of the country and even differentiate the impact of the toll road on the different regions of the country.

This experience tells us that there are major opportunities in converting Big Data into actionable information, but the impact of Big Data still remains limited. In our case, the use of mobile phone data helped generate timely and relatively inexpensive information on the impact of a large transport infrastructure on human mobility. On the other hand, it is clear that more analysis using socioeconomic data is needed to get to concrete and impactful policy actions. Thus, we think that making such information available to all stakeholders has the potential not only to guide policy action but also to spur it. 

References

Atkin, D. and D. Donaldson (2014). Who ’ s Getting Globalized ? The Size and Implications of Intranational Trade Costs . (February).

Clark, X., D. Dollar, and A. Micco (2004). Port efficiency, maritime transport costs, and bilateral trade. Journal of Development Economics 75(2), 417–450, December.

Donaldson, D. (2013). Railroads of the Raj: Estimating the Impact of Transportation Infrastructure. forthcoming, American Economic Review.

Fetzer Thiemo (2014) “Urban Road Construction and Human Commuting: Evidence from Dakar, Senegal.” Mimeo

Ji, Y. (2011). Understanding Human Mobility Patterns Through Mobile Phone Records : A cross-cultural Study.

Simini, F., M. C. Gonzalez, A. Maritan, and A.-L. Barab´asi (2012). A universal model for mobility and migration patterns. Nature 484(7392), 96–100, April.

Tinbergen, J. (1962). Shaping the World Economy; Suggestions for an International Economic Policy.

Yuan, Y. and M. Raubal (2013). Extracting dynamic urban mobility patterns from mobile phone data.

      

It is estimated that there are 95 mobile phone subscriptions per 100 inhabitants worldwide, and this boom has not been lost on the developing world, where the number of mobile users has also grown at rocket speed. In fact, in recent years the information communication technology (ICT) revolution has provided opportunities leading to “death of distance,” allowing many obstacles to better livelihoods, especially for those in remote regions, to disappear. Remarkably, though, the huge proportion of poverty-stricken populations in so many of those same regions persists.

How might, then, we think differently on the relationship between these two ideas? Can and how might ICTs act as an engine for eradicating poverty and improving the quality of life in terms of better livelihoods, strong education outcomes, and quality health? Do today's communication technologies hold such potential?

In particular, the mobile phone’s accessibility and use creates and provides us with an unprecedented volume of data on social interactions, mobility, and more. So, we ask: Can this data help us better understand, characterize, and alleviate poverty?

Mapping call data records, mobility, and economic activity

The first step towards alleviating poverty is to generate poverty maps. Currently, poverty maps are created using nationally representative household surveys, which require manpower and time. Such maps are generated at a coarse regional resolution and continue to lag for countries in sub-Saharan Africa compared to the rest of the world.

As call data records (CDRs) allow a view of the communication and mobility patterns of people at an unprecedented scale, we show how this data can be used to create much more detailed poverty maps efficiently and at a finer spatial resolution. Such maps will facilitate improved diagnosis of poverty and will assist public policy planners in initiating appropriate interventions, specifically at the decentralized level, to eradicate human poverty and ensure a higher quality of life.

How can we get such high resolution poverty maps from CDR data?

In order to create these detailed poverty maps, we first define the virtual network of a country as a “who-calls-whom” network. This signifies the macro-level view of connections or social ties between people, dissemination of information or knowledge, or dispersal of services. As calls are placed for a variety of reasons, including request for resources, information dissemination, personal etc., CDRs provide an interesting way to construct a virtual network for Senegal.

We start by quantifying the accessibility of mobile connectivity in Senegal, both spatially and across the population, using the CDR data. This quantification measures the amount of communication across various regions in Senegal. The result is a virtual network for Senegal, which is depicted in Figure 1. The circles in the map correspond to regional capitals, and the edges correspond to volume of mobile communication between them. Thicker edges mean higher volume of communication. Bigger circles mean heavier incoming and outgoing communication for that region.

Figure 1: Virtual network for Senegal with MPI as an overlay

Source: Author’s rendering of the virtual network of Senegal based on the dataset of CDRs provided as a part of D4D Senegal Challenge 2015

Figure 1 also shows the regional poverty index[1] as an overlay. A high poverty index corresponds to very poor regions, which are shown lighter green on the map. It is evident that regions with plenty of strong edges have lower poverty, while most poor regions appear isolated. 

Now, how can we give a more detailed look at the distribution of poverty? Using the virtual network, we extract quantitative metrics indicating the centrality of each region in Senegal. We then calculate centrality measures of all the arrondissements[2] within a region. We then correlate these regional centrality measures with the poverty index to build a regression model. Using the regression model, we predict the poverty index for each arrondissement.

Figure 2 shows the poverty map generated by our model for Senegal at an arrondissement level. It is interesting to see finer disaggregation of poverty to identify pockets of arrondissement, which are most in need of sustained growth. The poorer arrondissements are shown lighter green in color with high values for the poverty index.

Figure 2: Predicted poverty map at the arrondissement level for Senegal with MPI as an overlay

Source: Author’s rendering of the virtual network of Senegal based on the dataset of CDRs provided as a part of D4D Senegal Challenge 2015.

What is next for call data records and other Big Data in relation to eradicating poverty and improving the human development?

This investigation is only the beginning. Since poverty is a complex phenomenon, poverty maps showcasing multiple perspectives, such as ours, provide policymakers with better insights for effective responses for poverty eradication. As noted above, these maps can be used for decomposing information on deprivation of health, education, and living standards—the main indicators of human development index.

Even more particularly, we believe that this Big Data and our models can generate disaggregated poverty maps for Senegal based on gender, the urban/rural gap, or ethnic/social divisions. Such poverty maps will assist in policy planning for inclusive and sustained growth of all sections of society. Our methodology is generic and can be used to study other socio-economic indicators of the society.

Like many uses of Big Data, our model is in its nascent stages. Currently, we are working towards testing our methodology at the ground level in Senegal, so that it can be further updated based on the needs of the people and developmental interventions can be planned. The pilot project will help to "replicate" our methodology in other underdeveloped countries.

In the forthcoming post-2015 development agenda intergovernmental negotiations, the United Nations would like to ensure the “measurability, achievability of the targets” along with identification of 'technically rigorous indicators' for development. It is in this context that Big Data can be extremely helpful in tackling extreme poverty.

Note: This examination was part of the "Data for Development Senegal" Challenge, which focused on how to use Big Data for grass-root development. We took part in the Data Challenge, which was held in conjunction with NetMob 2015 at MIT from April 7-10, 2015. Our team received the National Statistics prize for our project titled, "Virtual Network and Poverty Analysis in Senegal.” This blog reflects the views of the authors only and does not reflect the views of the Africa Growth Initiative.

---

[1] As a measure of poverty, we have used the Multidimensional Poverty Index (MPI), which is a composite of 10 indicators across the three areas: education (years of schooling, school enrollment), health (malnutrition, child mortality), and living conditions.

[2] Senegal is divided into 14 administrative regions, which are further divided into 123 arrondissements.

      

Last month, we experienced yet another breakthrough in the epic battle of man against machine. Google’s AlphaGo won against the reigning Go champion Lee Sedol. This success, however, was different than that of IBM’s Deep Blue against Gary Kasparov in 1987. While Deep Blue still applied “brute force” to calculate all possible options ahead, AlphaGo was learning as the game progressed. And through this computing breakthrough that we can learn how to better leverage the data revolution.

In the game of Go, brute-force strategies don’t help because the total number of possible combinations exceeds the number of atoms in the universe. Some games, including some we played since childhood, were immune to computing “firepower” for a long time. For example, Connect Four wasn’t solved until 1995 with the conclusion being the first player can force a win. And checkers wasn’t until 2007, when Jonathan Schaeffer determined that in a perfect game, both sides could force a draw. For chess, a safe strategy has yet to be developed, meaning that we don’t know yet if white could force a win or, like in checkers, black could manage to hold on to a draw.

But most real-life situations are more complicated than chess, precisely because the universe of options is unlimited and solving them requires learning. If computers are to help, beyond their use as glorified calculators, they need to be able to learn. This is the starting point of the artificial intelligence movement.  In a world where perfection is impossible, you need well-informed intuition in order to advance. The first breakthrough in this space occurred when IBM’s Watson beat America’s Jeopardy! champions in 2011. These new intelligent machines operate in probabilities, not in certainty.

That being said, perfection remains important, especially when it comes to matters of life and death such as flying airplanes, constructing houses, or conducting heart surgery, as these areas require as much attention to detail as possible. At the same time, in many realms of life and policymaking we fall into a perfection trap. We often generate obsolete knowledge by attempting to explain things perfectly, when effective problem solving would have been better served by real-time estimates. We strive for exactitude when rough results, more often than not, are good enough.

By contrast, some of today’s breakthroughs are based on approximation. Think of Google Translate and Google’s search engine itself. The results are typically quite bad, but compared to the alternative of not having them at all, or spending hours leafing through an encyclopedia, they are wonderful. Moreover, once these imperfect breakthroughs are available, one can improve them iteratively. Only once the first IBM and Apple PCs were put on the market in the 1980s did the cycle of upgrading start, which still continues today.

In the realm of social and economic data, we have yet to reach this stage of “managed imperfection” and continuous upgrading. We are producing social and economic forecasts with solid 20th century methods. With extreme care we conduct poverty assessments and maps, usually taking at least a year to produce as they involve hundreds of enumerators, lengthy interviews and laborious data entry. Through these methods we are able to perfectly explain past events, but we fail to estimate current trends—even imperfectly.

The paradox of today’s big data era is that most of that data is poor and messy, even though the possibilities for improving it are unlimited. Almost every report from development institutions starts with a disclaimer highlighting “severe data limitations.” This is because only 0.5 percent of all the available data is actually being curated to be made usable. If data is the oil of the 21st century, we need data refineries to convert the raw product into something that can be consumed by the average person.

Thanks to the prevalence of mobile device and rapid advances in satellite technology, it is possible to produce more data faster, better, and cheaper. High-frequency data also makes it possible to make big data personal, which also increases the likelihood that people act on it. Ultimately, the breakthroughs in big data for development will be driven by managerial cultures, as has been the case with other successful ventures. Risk averse cultures pay great attention to perfection. They nurture the fear of mistakes and losing. Modern management accepts failure, encourages trial and error, and reaches progress through interaction and continuous upgrading.

       

DC KIDS COUNT, housed at the nonprofit DC Action for Children, is the DC chapter of a nationwide network of local-level organizations aiming to provide a community-by-community picture of the conditions of children. The 26 year-old project is funded by the Annie E. Casey Foundation and its aim is to provide high-quality data and trend analysis as well as help local governments monitor budget and legislative decisions based on evidence of what works for children and families. As we pointed out in our recent papers and a blog, developing reliable and comprehensive data is a critical step to building effective community partnerships and producing outcomes that improve economic mobility and health in a neighborhood.

We discussed these issues with HyeSook Chung, Executive Director of DC Action for Children.

Q. Please summarize the history of the DC Kids Count project. What motivated it, and how it has evolved over the last years?

A. As part of the nationwide Kids Count network, each chapter tracks a number of indicators on child and family well-being through an online database called Kids Count Data Center. Each chapter also releases a yearly data book which summarizes the state of child well-being within their state or locality. When DC Action for Children became the host of DC Kids Count in 2012, I wanted to rethink the way we presented our data to move beyond the traditional print format into the exciting realm of visualizing data. This led to the beginning of our partnership with DataKind, a group of dedicated pro-bono data scientists who worked with us to create an interactive, web-based data tool that maps out indicators of child well-being across DC’s 39 neighborhood clusters.

We know that the neighborhood children grow up in, and the resources they have access to, plays a huge role in shaping children’s future opportunities. The maps we created with our Data Tool 2.0  reveal sharp disparities in DC neighborhoods: some DC neighborhoods are wealthy and have many assets, while others are characterized by high levels of poverty. The many challenges that come with high poverty neighborhoods include: poorer performing schools, more crime, and less access to libraries, parks, and healthy foods.  

Q. What type of indicators do you gather? How many years does the data cover? What level of granularity does the data have?

A. We track a variety of indicators of child well-being, including demographics, economic well-being, health and safety. The data is housed online in two places: The KIDS COUNT Data Center and our Data Tool 2.0. The Data Tool 2.0 maps the most recent available data at the neighborhood cluster, while the Data Center allows for a wider range of geographies (citywide and ward level) and different timeframes.  Many of the indicators have data from 1990 to the present.

Q. How do you measure the data tool’s impact on policy and legislation?

A. We have made it a priority to conduct internal evaluations to assess the utilization of the online tool, but we also believe that measuring the tool’s impact must go beyond traditional web analytics. We regularly use the Data Tool 2.0 in our work with city officials and direct service providers to offer an overview of the social context in the city’s different neighborhoods.     

In a city where the allocation of resources is often guided by personal relationships and old-school politics, it is important to show clearly whether budget decisions are aligned with the needs of our children. We believe that our Data Tool 2.0 project can bring much needed transparency to the allocation of the DC government budget and help achieve agreement.

Q. The DC Kids Count project is helping build data capacity across organizations, with the aim of creating a more “data-literate” city. Could you tell us about some of these initiatives? 

A. Businesses like Amazon and Netflix increasingly focus on finding “actionable” insights from their data. For them, “big data” analytics can help answer tough business questions. With the right platforms for analytics, they can increase efficiency or even improve operations and sales.

In a similar manner, we at DC Action for Children believe that big data opens up the opportunity for us to improve and reshape our strategy and decision making process to better align services with the needs of DC children in the same way Amazon or Netflix does with their customers.

For instance, we are offering the Child and Family Services Agency technical and data analysis support for their Healthy Families Thriving Communities Collaboratives, which are a citywide network of community-based organizations designed to embed family supports in their communities. Their mission is to strengthen and stabilize families and to prevent child abuse and neglect by offering services in the form of case management and support. We use KIDS COUNT data at the ward and neighborhood levels to highlight needs in the community and inform their planning. This encourages the Collaboratives’ staff to look at data differently—integrating it as a vital part of their program planning and strategy.

Q. What are some of the obstacles and challenges you face in integrating the data, and updating it?

A. Historically, our data analysis looked at more traditional indicators, such as program enrollment and the number of child welfare cases. But now we think we can use our access to big data to pull out patterns within our datasets and help guide the decisions of the city administrators. For example, if we are trying to prevent future child abuse cases, we can look at patterns analyzing family and child data in specific neighborhoods. We can use the type of predictive analysis practiced in the for-profit business to help us serve DC children more efficiently and effectively.

One of the most significant obstacles we face is ensuring that the indicators are up-to-date. This can be an issue with government agencies since some of them are slow in their release of new data.  Moreover, there is also no standard format across local agencies for how data is collected and released. Furthermore, data is often aggregated at different geographical units, like zip codes or census tracts. To get the data ready to upload to our Data Tool, we must recalculate the data into neighborhood clusters.  

Q. What policy changes would help produce better data-sharing ecosystems? 

A. DC has in many ways demonstrated leadership in data sharing. The Office of the Chief Technology Officer works to make a large variety of datasets publicly available. We have also seen large investments over the years to create new data systems that track progress and service delivery for different agencies. But our city can do more to promote a data-sharing ecosystem. So can other cities.

While multiple agencies are adopting innovative data systems, the systems are often siloed and do not speak to each other. Moreover, since data is tracked differently across agencies, based on needs and requirements for reporting, it is difficult for agencies to share data both publicly and internally. It is also often difficult to get access to de-identified disaggregated data for richer analysis. We are glad that many agencies recognize the value of robust data collection, but more data transparency policies would give us a better understanding of the challenges that lie behind improving the wellbeing of children in the city.

Q. What are the next steps for the DC Kids Count project, and how do you expect it to grow over the next few years?

A. We just finished wrapping up some of the final work on our DataTool 2.0. In terms of next steps, we are working on a handbook that explains how we created our Data Tool so that other Kids Count chapters and organizations can replicate and adapt our tool.

We would also like to add local budget data to the asset maps to see if public investments align with the neighborhoods that need it the most. This would give us a more nuanced understanding of the geography of DC budget investments, including inequities in investments by geography and demographics.

Big data analytics has changed the way we focus our priorities and engage in business practices. I’m committed to this movement. I think that, through big data, we can also revolutionize the way we do policy.

***

In conclusion, DC Kids Count, housed at the nonprofit DC Action for Children, belongs to a larger, nationwide group of organizations helping to better coordinate regional development through data-driven decision making. By centralizing different government databases, and providing real-time, community level data, DC Kids Count can help local government entities allocate their resources more efficiently and creatively and help foster place-conscious strategies. The process behind compiling the data also illustrates many of the challenges—data sharing, interoperability of data systems, access to real-time data involved in building “data- sharing ecosystems.”

      

There are lots of good reasons to go to college, but the vast majority of prospective students in this country report[i] that they’ll go to college because they believe that it will improve their employment opportunities and financial wellbeing. And for the most part, they’re right. Despite many suggestions to the contrary, it’s very well documented[ii] that investments in higher education pay large dividends in the form of future earnings. This makes higher education one of the most important tools we have for generating social mobility. Regardless of an individual’s starting point in life, higher education offers access to greater financial well-being. Unfortunately, it’s not a fail proof system.

Investments in education, like investments in the stock market, do not come without risk. In financial markets, access to information is one way investors mitigate risk. Mutual funds, for example, disclose average returns over various time periods for certain categories of investments (e.g. large-cap funds, emerging market funds, technology funds, etc.), in addition to other information. These data, moreover, are widely and freely available through consumer-oriented websites like Yahoo Finance, Vanguard, and E-Trade. Yet, for higher education, students have had access to no analogous information until quite recently.

For decades, economists discussed the average benefits of a college education compared to a high school education with no regard to either field of study or institution. Finally, in 2009, the Census Bureau started collecting data that could be used to assess which majors pay the most,[iii] and then just a few months ago, the Department of Education released data on the earnings of alumni by institution, for all students who receive federal grants or loans. These data can be further analyzed, as we have done, to estimate the economic contribution of schools (or value-added) as distinct from the outcomes attributable to student characteristics (like test scores).[iv] Still, even with these data advances, students cannot compare earnings by major across institutions, except in a handful of cases using state data systems.

Here, we illustrate how data by major and institution can inform the decision of what to study and where using data from Texas. Suppose first that this student is a Texas resident and has decided she would like to pursue a bachelor’s degree at a public institution in her state.

Our data on alumni earnings by major comes from the Texas Higher Education Board, and we combine it with information on the net cost of tuition from the Department of Education’s IPEDS database as reported in the College Scorecard.[v] We use these data to estimate the ten-year return on investment for each institution in the state of Texas by major.   

We calculate an estimate of ten-year return by summing the average earnings faced by graduates over the first ten years following graduation[vi] and subtracting off the wage they would have received as a high school graduate without a degree (taking into account additional years of earnings when they would have been enrolled in college). To estimate this benchmark, we used data on Texas residents from the Annual Social and Economic Supplement to the Current Population Survey, obtained via IPUMS CPS.[vii] We then subtract the institution specific costs[viii] to get the ten-year financial return. Since education pays off over a lifetime, this isn’t the ideal exercise, but it’s still informative. We’ve estimated these returns based on the population of individuals who both complete their degree and do not go on to complete graduate study. Ideally, these estimated expected returns would be adjusted to account for how earnings and costs are affected by non-completion. Indeed, the average rate of completion across these schools is only 48 percent. This is a quick and dirty method for estimating returns that fails to take into account a number of selection issues,[ix] but we believe that it still provides an effective illustration of risk in higher education. 

Figure 1 illustrates the potential average outcome facing our Texas student, who is deciding between bachelor’s degree programs from the set of public institutions in her home state. We’ve plotted the distribution of financial returns for the set of potential expected outcomes, which are defined as all combinations of institution and major. To be clear, the distribution of potential outcomes would be far wider if we were using individual specific variation (i.e. the fact that some students will ultimately earn more than others, even with the same degree from the same institution) and the real possibility of non-completion.

We know that, on average, this student will face a positive return on her investment, wherever she chooses to go. The average rate of return across all possible choices facing this student is quite a sizeable 11.3 percent (or $216,000 in undiscounted 2014 dollars). At a systemic level that’s important.  Still, the standard deviation is 6.7, with a low return of a -6.6 percent (Animal Science at Sul Ross State) and a high return of 79.8 percent (Registered Nursing at UT Brownsville). Out of 1065 combinations of majors and schools, 19 yielded average negative returns. This was true even for two programs at the selective UT Austin campus (Visual and Performing Arts and Classics). 1.1 percent of students who graduated in 2004 were in a major-institution combination that yielded a net return below 4 percent. In such cases, they would have been better off putting their dollars into treasury bills.

Figure 1. Mean return on bachelor’s degree investment by institution and major, for Texas residents who graduated in 2004 from a Texas public college

Students who know what they want to major in could benefit greatly from knowing which school is likely to generate the largest pay off (it would be nice to know this in terms of learning as well as money, but that is another more complicated matter).

We’ve illustrated the distribution of potential outcomes for two different popular majors, Liberal Arts and Sciences and Electrical Engineering.[x] Both majors clearly offer a significant average rate of return across all institutions (12 for Liberal Arts and 20 for Electrical Engineering), but depending on which major they choose the student will face a different level of risk in their future earnings. The variation (standard deviation) in the expected rate of return across institutions is much larger for Liberal Arts majors (5.7) than for Electrical Engineering majors (3.7). Yet, while these facts may discourage people from pursuing a Liberal Arts major in the abstract, the plot below does show that some Liberal Arts majors out-earn their peers in electrical engineering. For example, Liberal Arts majors from UT Austin earned a higher return than electrical engineering majors at UT Dallas, the University of Houston, and three other UT campuses. Thus, these more detailed facts can actually encourage students to pursue majors that look economically bad for the average student but quite attractive at a particular school with a strong program.

Figure 2. Distribution of earnings 10 years after graduation for bachelor’s degree holders with an Electrical Engineering or Liberal Arts degree, for Texas residents and 2004 graduates from Texas public colleges 

The point is that college degrees, like other investments, are risky, but information goes a long way to clarify the nature of that risk and improve the quality of investment decisions. In addition to providing students and the public greater access to data on market performance of alumni, there are a number of innovations both in the policy arena and in the private market that could help make college investments less risky. First of all, innovative financing systems that allow students to pay for their investment over a longer period of time and tie repayment to earnings would greatly limit downside risk for students.  Second, institutions have the capacity to shoulder some of this risk, and a proposal known as risk-sharing[xi] is gaining some traction and would require schools to pay the federal government some portion of loan default losses. On a voluntary basis, some colleges have offered on-time graduation guarantees[xii] and wage guarantees.[xiii] And last, new business models in higher education could help mitigate risk. Part of the problem in the current system comes from the all-or-nothing regime in which students have to invest in a bundle of coursework (i.e. a degree) in order to reap significant returns. The growing prominence of new models, like micro-credentials[xiv] and coding boot camps,[xv] can offer alternatives that don’t require students to put all of their eggs in one basket.

       

Each year millions of low- to moderate-income Americans supplement their income by claiming the Earned Income Tax Credit (EITC) during tax season. Last year, 1 in 5 taxpayers claimed the credit and earned an average of nearly $2,400.

This tax season, some of those eligible for the EITC may also be able to claim, for the first time, a new credit created by the Affordable Care Act (ACA) to offset the cost of purchasing health insurance for lower-income Americans. It’s called the ACA premium tax credit.

To qualify for the ACA premium tax credit, filers need first to have an annual income that falls between 100 and 400 percent of the federal poverty line (between $11,670 and $46,680 for a single-person household in 2014). Beyond the income requirements, however, filers must also be ineligible for other public or private insurance options like Medicaid or an employer-provided plan.

Why the tax credit overlap matters

Identifying the Americans eligible for both credits is important because it sheds light on how many still need help paying for health insurance even after the ACA extended coverage options.

In a recent study of the EITC-eligible population, Elizabeth Kneebone, Jane R. Williams, and Natalie Holmes estimated what share of EITC-eligible filers might also qualify for the ACA premium tax credit this year.

Below, see a list of the top 10 states with the largest overlap between filers eligible for the EITC and those estimated to qualify for the ACA premium tax credit.* Notably, none of these states has expanded Medicaid coverage to low-income families after the passage of the ACA.

Nationally, an estimated 7.5 million people (4.2 million “tax units”) are likely eligible for both the ACA premium tax credit and the EITC. Nearly 1.3 million of those tax units are from the following ten states.

1. Florida

Overlap: 22.5 percent / 405,924 tax units
State-based exchange? No Expanded Medicaid coverage? No

2. Texas

Overlap: 21.4 percent / 513,061 tax units
State-based exchange? No Expanded Medicaid coverage? No

3. South Dakota

Overlap: 20.5 percent / 15,124 tax units
State-based exchange? No Expanded Medicaid coverage? No

4. Georgia

Overlap: 19.8 percent / 186,020 tax units
State-based exchange? No Expanded Medicaid coverage? No

5. Louisiana

Overlap: 19.6 percent / 86,512 tax units
State-based exchange? No Expanded Medicaid coverage? No

6. Idaho

Overlap: 19.3 percent / 28,855 tax units
State-based exchange? Yes Expanded Medicaid coverage? No

7. Montana

Overlap: 18.9 percent / 18,138 tax units
State-based exchange? No Expanded Medicaid coverage? No

8. Wyoming

Overlap: 18.4 percent / 7,276 tax units
State-based exchange? No Expanded Medicaid coverage? No

9. Utah

Overlap: 18.1 percent / 42,284
State-based exchange? No (Utah runs a small businesses marketplace, but it relies on the federal government for an individual marketplace) Expanded Medicaid coverage? No

10. Oklahoma

Overlap: 18.0% / 63,045 tax units
State-based exchange? No Expanded Medicaid coverage? No

* For the purposes of this list, we measured the overlap in “tax units,” not people. One tax unit equals a single tax return. If a family of four together qualifies for the ACA premium tax credit, they would be counted as one tax unit, not four, since they filed jointly with one tax return.

       

One in five tax filers in the United States claims the Earned Income Tax Credit—a refundable federal tax credit targeted to low-income working Americans that has proven to be one of the nation’s most effective anti-poverty policies. Last year, at tax time the average EITC filer claimed just over $2,400 through the credit.

However, the share of filers claiming the EITC and the level of benefits they receive vary widely within and across communities, as shown by the local-level IRS data we post each year on our EITC Interactive data tool. For instance, almost one in three filers in the Memphis metro area claimed the credit (32 percent) in tax year 2013 compared to just 12 percent of filers in metro Boston. Local labor market conditions can affect these numbers, like the incidence and concentration of low-wage jobs or regional differences in cost of living and average wage levels. But the credit itself is also designed to vary across different kinds of filers and families. Maximum credit levels for workers without children are quite small, but they increase considerably for workers with one, two, or three children—boosting the credit’s work incentive and anti-poverty impacts.

For the first time, our EITC Interactive tool now includes data on how EITC receipt varies by the number of children claimed.

According to that data, last tax year workers without qualifying children received an average credit of $281 (Figure 1). Although they made up almost one in four EITC filers, childless workers accounted for just 3 percent of EITC dollars claimed, due to the small size of their credit (Figure 2). In contrast, workers with one child—the largest share of EITC filers (37 percent)—claimed an average credit of $2,316. Workers with two kids accounted for 27 percent of EITC filers, but with an average credit of $3,682 they took home 40 percent of all EITC dollars. Working families with three or more children made up the smallest share of EITC filers last tax year, but claimed the largest credit on average at $4,036.

These data, which are available down to the ZIP code level, offer insights into the ways in which the makeup of the EITC population (and the low-wage workforce more generally) varies across places. Returning to the Memphis and Boston regions, each metro area received more than half a billion dollars through the EITC last year ($517 and $512 million, respectively). However, the number of filers claiming the EITC was much larger in metro Boston (256,456) than in the Memphis metro area (178,241). In part, these numbers reflect the fact that 30 percent of metro Boston’s EITC filers were childless workers. In the Memphis metro area, just 15 percent of EITC filers did not have qualifying children, while 41 percent had one child, 31 percent had two children, and 12 percent had three or more children—higher than Boston’s share of EITC filers with children across the board (37 percent had one child, 24 percent had two children, and 9 percent had three or more children).

For EITC outreach campaigns working to ensure eligible filers claim the EITC at tax time, and for practitioners looking to use tax time to connect low-income workers to financial services and benefits, these numbers give a sense of who lives in their community and how to target their services. For advocates and policymakers, these numbers help shed light on how potential changes to the credit might affect different places.

For instance, the Obama administration, several legislators, and at least one presidential candidate have proposed expanding the EITC for workers without qualifying children to make it a more effective poverty alleviation and work support tool. Every congressional district in the country has childless workers or noncustodial parents who would stand to benefit from that expansion. But that expansion would be particularly important for the more than 240 districts—largely clustered on the coasts and roughly split between Republican and Democratic representatives—with above average shares of childless EITC filers (Map 1).

In contrast, if Congress does not act to make recent expansions to the credit permanent, every district will see a cut in EITC benefits in 2017, when the credit for workers with three or more children is set to disappear. In particular, more than 200 districts with above average shares of EITC filers with three or more kids—this time predominantly Republican districts clustered in the Intermountain West, parts of the Great Plains, and along the Texas border—would be most affected (Map 2). 

In the coming weeks, we will be delving deeper into the impact of proposed and potential changes to the EITC and releasing new resources on the EITC-eligible population and the credit’s anti-poverty impact. In the meantime, these new EITC Interactive data offer an important resource that can help practitioners, policymakers, advocates, and researchers better understand how the EITC affects low-income workers and families and their communities across the country.

       

       

       

Event Information

Strategies for facilitating biomarker development

The emerging field of precision medicine continues to offer hope for improving patient outcomes and accelerating the development of innovative and effective therapies that are tailored to the unique characteristics of each patient. To date, however, progress in the development of precision medicines has been limited due to a lack of reliable biomarkers for many diseases. Biomarkers include any defined characteristic—ranging from blood pressure to gene mutations—that can be used to measure normal biological processes, disease processes, or responses to an exposure or intervention. They can be extremely powerful tools for guiding decision-making in both drug development and clinical practice, but developing enough scientific evidence to support their use requires substantial time and resources, and there are many scientific, regulatory, and logistical challenges that impede progress in this area.

On October 27th, 2015, the Center for Health Policy at The Brookings Institution convened an expert workshop that included leaders from government, industry, academia, and patient advocacy groups to identify and discuss strategies for addressing these challenges. Discussion focused on several key areas: the development of a universal language for biomarker development, strategies for increasing clarity on the various pathways for biomarker development and regulatory acceptance, and approaches to improving collaboration and alignment among the various groups involved in biomarker development, including strategies for increasing data standardization and sharing. The workshop generated numerous policy recommendations for a more cohesive national plan of action to advance precision medicine.  

Event Materials

• 1027 Brookings biomarkers workshop agenda
• 1027 Biomarkers workshop backgrounderfinal
• 1027 Biomarkers workshop slide deckfinal
• 1027 Biomarkers workshop participant listfinal