TooEarly publishes applied economics research using data analysis and modern computational methods, from food security to labor markets to policy effectiveness.
Why "TooEarly"?
In diplomatic folklore, there's a famous exchange: Henry Kissinger asks Chinese Premier Zhou Enlai what he thinks of the French Revolution. Zhou pauses and replies, "It is too early to say."
The story became legendary as evidence of Chinese long-term thinking—measuring history in centuries while the West fixates on election cycles. But the myth obscures a more interesting reality. Zhou wasn't talking about 1789. Zhou's interpreter, Chas Freeman, later clarified: Zhou had been referring to the Paris student uprisings of 1968, which had happened just three years earlier.
The real lesson is about communication and evidence. Zhou gave a reasonable answer about events still unfolding. His audience heard what they expected to hear. Decades passed before anyone checked the facts.
In economics, similar dynamics play out constantly. Policymakers demand immediate verdicts on complex interventions. Stakeholders cherry-pick preliminary data to support predetermined conclusions. Researchers face pressure to publish findings before mechanisms are clear.
When evidence is insufficient, we should acknowledge uncertainty. When data quality matters for conclusions, we document our validation process. When causal mechanisms remain unclear, we distinguish what we know from what we don't. And when evidence speaks clearly—whether three years or thirty years after an event—we communicate findings in language that policymakers and communities can use.
What We Research
Here, we'll discuss several areas:
Applied Economics Research
We analyze real-world economic questions using administrative data and modern computational methods. Our research examines food security vulnerability, labor market dynamics, policy intervention effectiveness, and the economic barriers facing working families. Each analysis uses public data sources and transparent methodology to answer questions that matter for communities and policymakers.
Methodological Transparency
We document how we do research, including the mistakes we catch and the decisions we make along the way. Our methodology posts walk through computational approaches (calculating 2.7 million transit routes to measure food access), data validation processes (verifying 6,613 grocery store locations), and causal inference techniques (using difference-in-differences to isolate policy effects). We share code, explain our reasoning, and provide replication materials.
AI-Integrated Workflows
We incorporate AI tools throughout the research process while maintaining rigorous standards. Our workflows demonstrate how large language models can assist while keeping human judgment central to analysis and interpretation. We share our AI-augmented research processes to help other researchers work more efficiently.
Policy-Relevant Insights
Our research quantifies disparities, tests conventional wisdom, and translates findings into actionable recommendations. We analyze SNAP program effectiveness, document the "working poor" phenomenon, examine housing costs as food policy, measure transit accessibility barriers, and assess differential impacts of economic shocks across vulnerable communities.
Our Approach
Evidence guides conclusions. We follow the data wherever it leads, document our methods transparently, and acknowledge uncertainty when appropriate. Our analyses use public administrative data from sources like the Census Bureau, state SNAP programs, and the Bureau of Labor Statistics.
Rigorous and accessible. We maintain the same analytical standards as peer-reviewed journals while explaining findings in plain language. Clear communication strengthens research impact.
Open methods and materials. We share code, document validation processes, and provide replication materials. When we catch errors or refine our approach, we explain what changed and why. This transparency helps other researchers and strengthens confidence in findings.
Timely and actionable. We focus on questions that matter for communities and policymakers right now. Our research translates into specific recommendations grounded in empirical evidence, delivered when decisions need to be made.
Who This Is For Anyone
-
Seeking evidence-based analysis to inform decision-making on economic policy questions.
-
Interested in applied methods and replication materials for causal inference studies.
-
Covering economic policy who need rigorous sources and clear explanations of complex issues.
-
Who want depth and nuance, not soundbites and oversimplification.
Research Standards
We hold ourselves to high standards:
- Public data sources: Census Bureau, SNAP administrative records, Bureau of Labor Statistics, and other government datasets
- Peer-reviewed methods: Difference-in-differences, synthetic control, causal forests, and other established econometric techniques
- Replicable analysis: Full methodology documentation and code available on request
- Proper attribution: Every claim is cited to its source; every dataset is documented
- Honest about limitations: We acknowledge caveats, uncertainty, and cases where the evidence is insufficient
Get Involved
TooEarly is dedicated to rigorous, accessible applied economics research. We value reader feedback and questions. Posts may includes discussion prompts and a response form. We read every submission and use reader insights to identify new research questions.
For researchers seeking replication materials, methodology details, or data documentation, contact us through the site. We're committed to making our work reproducible and transparent.
Subscribe to receive new analysis when we publish.
Explore our research to see our approach in action.
Share your perspective on the questions we're investigating.