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

Google open sources differential privacy tool. Google recently open sourced a version of their differential privacy algorithm, which is used in many of its core products, to help government officials and business owners gather useful information from sensitive data while ensuring the privacy of the people behind the data. Link and link

Tax-Calculator analysis in AEI tax blog. AEI’s Aparna Mathur launched a new blog series that features a range of perspectives on the Tax Cuts and Jobs Act (TCJA) of 2017. As part of the blog series, AEI’s Erin Melly will publish a series of figures to visualize the effects of the TCJA. In the first of her series, Melly uses Tax-Calculator* to show the TJCA’s distributional effect on after-tax income. Link

NYU law professors’ new paper on economic inequality and fiscal policy. In a recent paper titled “Taxing the Rich: Issues and Options,” Lily Batchelder and David Kamin explore strategies for raising tax revenues to reduce economic inequality in America. Among the various policy options they explore, Batchelder and Kamin use the Tax-Brain* web application to estimate the revenue generated from raising the top income tax rate from 37 percent to 39.6 percent. To generate their estimate, the authors use Tax-Brain’s behavioral response capabilities to account for individuals’ responses to an increased tax rate. A link to their Tax-Brain simulation can be found here. The paper can be downloaded here.

Open source analysis of Bernie Sanders’ new wealth tax proposal. Emmanuel Saez and Gabriel Zucman, the economists who assessed Elizabeth Warren’s wealth tax proposal (see our April newsletter for more), use their open source wealth tax calculator to estimate the effects of Sanders’ progressive wealth tax on the top 0.1 percent of Americans by wealth. Their online application allows its users to guess the tax evasion percentage for themselves, and the source code is available for those who would like to adjust other assumptions. Link

Transparency and reproducibility at the Policy Simulation Library (PSL) meeting. At the September PSL* meeting, Lars Vilhuber, a senior research associate at Cornell University and data editor for the American Economic Association (AEA), discussed the history of, challenges to, and progress toward replicability in the social sciences and the concrete steps that the AEA has taken to improve data and code transparency in its journals. Link

Policy Change Index (PCI) at the Fed. Last week, Julian TszKin Chan (Bates White Economic Consulting) and Weifeng Zhong (Mercatus Center) presented PCI-China,* a tool for predicting Chinese public policy with machine learning, to staff at the Federal Reserve Board. Link

* These projects are attendees or graduates of OSPC’s incubator program.

Edited by Matt Jensen and Peter Metz