Making Cities Work

Cities are the economic engines of the world, but they face huge challenges, including congestion, pollution, spiraling housing costs, and sprawl that outpaces urban infrastructure, especially in low- and middle-income countries (LMICs). Excessive regulation, restrictive zoning, and constraints on development further exacerbate these problems. The well-being of billions depends on thriving cities, and thus on smart urban policy and planning.

Cities are notoriously bad at learning lessons from each other, especially across countries. In part, this is because of a lack of data that makes it hard to compare which policies have worked for whom. We are building an open data platform and a research agenda documenting the dramatic and fixable gaps in city performance around the world.

This project is analogous to the Penn World Tables, which were the first broadly comparable cross-country national accounts data at global scale. They were the foundation of a generation of economic research. They shaped what questions could be asked and answered in development and macroeconomics, and became a standard reference that influenced both frontier research and policy debates around the world for decades. Similarly, we aim to build a global standard for economic data on cities in the developed and developing world.

For historical reasons, statistical agencies in LMICs have heavy rural biases. Researchers have followed the data, leaving cities under-studied even as they grew to dominate the economic landscapes of their countries. An abundance of digital data and new AI tools make it possible to get data on cities at scale and speeds previously unimaginable. We are aggregating four kinds of data:

• Satellite data and global research outputs from Google, Meta, etc.;
• Regulations and policies extracted from municipal documents with LLMs;
• Subnational surveys (e.g., labor force, health, values) which can be aggregated to cities with some work;
• Novel surveys of experts, developers, and local governments in the style of the Doing Business surveys.

We are aiming for a big data release in Fall 2026. For now, you can read about our preliminary work creating a standard set of open, functional boundaries for cities around the world.