Nominal News is an economics newsletter written by a PhD Economist that translates the latest economic research into clear, policy‑relevant insights on current issues. Join 4,000 readers to stay-up-to-date with Nominal News directly in your inbox:
Quick update from Nominal News: we have created a list of Economics newsletters/writers on Substack. This list also allows you to track these individuals in the Notes feed https://www.nominalnews.com/lists/econstack:
Today’s article will be a bit different, as I will discuss an important econometric/statistical assumption that underlies all causal data analysis called SUTVA (Stable Unit Treatment Value Assumption). The importance of this assumption in data analysis is paramount - if SUTVA is heavily violated, any causal analysis is likely to not be valid and can’t be relied on. It is important to be aware of this assumption, and whether it is satisfied, when observing data analysis in the media or conducting data analysis in a work setting.
The inspiration for this post is the ‘analyses’ that I saw published after the recent U.S. midterm elections. Right after the election on November 3, 2026, a discussion erupted on which type of Democratic candidate (progressive, liberal, moderate, conservative, etc.)1 can get better election results - i.e. what is the causal impact of being a ‘moderate’ candidate on election outcomes. Below are a few examples of such analyses:
How Leftists and Moderates Did in the Midterms by Jeff Maurer
Strength in Numbers by G Elliot Morris
The problem - none of the above analyses are likely to be valid.
