What should financial regulators do about systemic risk? Ideally, many would like to describe, track, and aggregate the implications of nearly every significant financial contract around the world. Though daunting in scope, doing so would be technically quite feasible. The coordination necessary to make such a system work would be much more challenging than building it. Indeed, at smaller scales, professional risk managers already describe, track, and aggregate contract implications every day. Their data systems, however, are ad hoc and proprietary. Both the inputs and outputs of their risk calculations may be totally incomparable across different organizations-or even within the same institution. What's needed is a way to standardize the characterization of financial contracts. An international team led by the Stevens Institute of Technology is already working on the open-source software needed. They claim that the cash flow implications of nearly any financial agreement can be accurately approximated using just 30 standardized "contract types." Like Lego blocks, these can fit together to model quite complicated and comprehensive structures. Their widespread use would give both regulators and financial institutions the ability to "put all the pieces together" and model financial risk in ways that are impossible now. This grant provides funds to support the development and deployment of a pilot open-source "contract typing" software system with the ability to accurately model the cash flow implications of a wide range of financial contracts. Funds will support software development, testing, refinement, infrastructure, and outreach in an attempt to demonstrate the feasibility of such a software system.

Funds from this grant support a series of data-oriented improvements to the influential Brookings Papers on Economic Activity (BPEA) that will make its papers as replicable, interactive, and technologically empowered as those at the forefront of innovation in other fields. Planned improvements include adding the capacity to include new data as they arrive, thus keeping the analyses and conclusions in a paper up-to-date for years after publication. Also planned are upgrades that will enable papers to come with embedded code and interactive data visualizations that allow readers to test alternative regression specifications, change parameter settings, or adjust the time frame analyzed, all within the paper. Funds would primarily support the requisite technical updates to the BPEA website, with additional monies for user support and for incentive funds for authors who commit to making periodic updates to their data and findings.

Founded 50 years ago, the Inter-University Consortium for Political and Social Research (ICPSR) provides leadership and training in data access, curation, and methods of analysis for the social science research community. Over 700 institutions from all over the world belong to this consortium based at the University of Michigan, and its archives contain over 500,000 data files. This grant funds a project led by economic historian and ICPSR Director George Alter to help set standards and address challenges common to social science researchers who work with "big data." Grant funds will support three workshops that will aim to (1) develop consensus among social science journal editors about how to review, publish, and cite data; (2) develop common standards in a variety of scientific fields about how to archive data files and the "metadata" that describes them; and (3) develop a consensus among scientific grantmaking organizations about what data management standards should be imposed on grantees. Additional monies from this grant support a project to investigate the nondisclosure agreements (NDA) many social scientists sign in order to gain access to proprietary information and to explore the possibility of developing a common non-disclosure agreement on the model of the popular license developed by Creative Commons.

One of the few advantages of our complex tax code is that the information gathered can, in principle, provide researchers with accurate estimates of wages, investments, retirement savings, and many other economic variables. In practice, however, it is very hard for researchers to gain access to that information. Recognizing the demand for such data, the Internal Revenue Service has begun making more of its information available in aggregated tables and in de-identified compilations known as "Public Use Files." This two-year grant funds a project by the Tax Policy Center (TPC), a joint venture of the Urban Institute and the Brookings Institution, to help make IRS data more useful to researchers, policymakers, and the public. Over the course of the next two years, researchers at the Tax Policy Center propose to add new information to existing IRS data offerings, including data about age, gender, and how joint earnings are split between couples. They will also develop new methodologies for estimating the characteristics of those who do not file taxes, allowing more robust conclusions to be drawn from IRS data. New data and methodologies will be developed and added in ways that protect taxpayer anonymity and privacy.

Casual discussions of the scientific and technical workforce often rely on a pipeline metaphor. In this picture, there is an ample supply of student interest to begin with, but leakage at critical junctures leaves only a trickle of graduates who actually pursue careers in STEM. The obvious remedy is to plug the leaks. But perhaps the "pipeline" theory is an easy but misleading oversimplification. This grant supports a project by Hal Salzman of Rutgers to investigate how various pathways can lead through the educational system to STEM careers. Using the Baccalaureate and Beyond Longitudinal Study (B&B) compiled by the National Center for Education Statistics, Salzman and his team will analyze the complex ways that course-taking patterns relate to decisions about STEM majors and careers, including how students (a) use college as a period of exploration; (b) may benefit from majoring in STEM without pursuing a traditional STEM career; (c) can major in a non-STEM field but still do lots of science in classes or at work; (d) make choices that are influenced by both supply and demand variables; and (e) can thereby end up in scientific careers by way of nonlinear and nontraditional routes. The resulting picture, complemented by a series of interviews with students and site views to universities, promises to help build a more robust, nuanced of the myriad ways in which students may end up in scientific careers.

Gross domestic product (GDP) is the most important statistic in macroeconomics. As a measure of the value of goods and services produced within a country, GDP announcements can swing stock markets, political sentiments, business plans, and much else. Yet despite its importance, GDP figures--calculated in the U.S. by the Bureau of Economic Analysis (BEA) --are merely estimates and often subject to substantial subsequent revision. With businesses, politicians, and consumers making choices based on GDP data, however, such revisions can be costly. With so much at stake, the methodology for estimating GDP and similar statistics is the subject of constant scrutiny. This grant supports a major international conference about macroeconomic statistics to be held in August 2012. Conference participants will include a host of venerable research and government institutions, including the BEA, the International Monetary Fund, the U.S. Census Bureau, the U.S. Bureau of Labor Statistics, and the National Bureau of Economic Research. Grant funds will offset conference costs, support the commissioning of papers for conference sessions on GDP revisions and new GDP data sources, and enable the publication of a selection of peer-reviewed papers from the conference.