This is a periodic newsletter of the interesting things we’ve seen and what we are thinking about in open source policy analysis.

A balancing act between transparency and privacy. For decades, social science researchers and government planners have relied on data from the Census Bureau to conduct research on the economy, demographics, and public health. However, the days of public access to the reliable Census Bureau data that we’ve grown to love may be numbered. In September, the Census Bureau announced that it would adopt a new, more rigid standard called differential privacy to protect individuals’ identities. The Census Bureau argues that this new standard is necessary because, despite current safeguards to protect individual information, person-level data can be reconstructed by coupling Census-reported data with data available from other sources. On the other hand, some social science researchers believe that the Census Bureau’s new methods are overly cautious and experimental and claim that this new standard will lock up data that is indispensable to research and policy analysis. Given that the Census Bureau is a role model for other federal, state, and local data and statistical agencies, it is almost certain that the differential privacy debate will soon spread beyond Census. Link and Link

Cost matters when evaluating Paid Family Leave policy proposals. A new survey from the Cato Institute on support for federal paid leave programs shows that Americans support paid leave proposals in negative proportion with the proposals’ estimated cost to the taxpayer. The survey uses low-, mid-, and high-range cost estimates of policy proposals based on the AEI-Brookings Working Group on Paid Family Leave’s Cost Calculator, which is an interface to the Paid Family Leave–Cost Model (PFL-CM), a project incubated by OSPC. Link

A new paradigm for scientific research in Europe. A group of research funders from 11 different European countries have unveiled their radical vision for the future of scientific publications. The initiative, called “Plan S,” requires scientists they fund to publish their work in open-access or freely accessible websites by 2020. This policy is a stark departure from present day, where some 85 percent of journal articles reside in subscription-based publications, such as Nature and Science. Open-access publishing is itself controversial, because it shifts the costs of publishing to scientists themselves, who must pay a fee to publish. Link

OG-USA presented at Congressional Budget Office symposium. In a recent overlapping generations model symposium hosted by the CBO, Richard Evans (University of Chicago) and Jason DeBacker (University of South Carolina) presented their work on OG-USA, an OSPC-incubated model. OG-USA is an overlapping generations model of the US economy for evaluating the dynamic effects of fiscal policy. Link

Spotlight on PyData. PyData is an educational program of NumFOCUS that brings together open source users, developers, and researchers in a range of industries – from aerospace to architecture to policy analysis – to educate and develop new analytical methods. This fall, OSPC’s Anderson Frailey presented on the emergence of open policy analysis at a PyData-hosted event in New York City. Link and Link

Edited by Matt Jensen
American Enterprise Institute