Reflection: Gold-standard data powers deeper understanding of incomes in America

When I first heard about the Income Distributions and Dynamics in America project, I was immediately reminded of the countless students who have walked into my office over the past decade. They often come with an interesting question, but they face a key constraint: lack of data. Sometimes they want to study economic outcomes for specific demographic groups that are too small to be represented in the American Community Survey or Current Population Survey. Other times, they want to study the impact of a trade shock or technology change not just on average incomes but on inequality and income volatility. Some students want to map incomes experienced by people today back to where those people lived a few years ago prior to when some policy began.

These data constraints mean my students often cannot answer the big questions that motivate them. They might want to know how incomes have grown for Native communities relative to U.S. earners as a whole, or if the move of a major employer to their home state changed income inequality over time. Publicly available data just aren't granular and detailed enough to answer these types of questions.

Now, the obvious solution to the problems these students face is to get access to tax data from the IRS or demographic data from the Census Bureau (or better yet: linked Census Bureau-IRS data). The challenge, of course, is that this is far from straightforward. The IRS and Census Bureau go to incredible lengths to support research, but they face constraints on their time. As a result, there are limited numbers of projects they can support.

The Income Distributions and Dynamics in America resource takes a giant leap in expanding access to the powerful information that is held in administrative data. Through a research collaboration with the Census Bureau, this resource makes available to everyone millions of statistics on income inequality and income mobility over the life cycle-broken down to the state level and into detailed demographic subgroups. The backbone for the data construction in these statistics is linked Census Bureau-IRS data: Form 1040s and Form W-2s are linked to the demographic information people provide on the decennial census and other records. I may be slightly biased, as I've had the pleasure to work with this individual-level data, but this backbone dataset is the gold standard for measuring incomes in the U.S. The statistics, in my view, are an exciting public good for every researcher, student, and organization seeking to better understand the income experiences of earners in America.

This article is featured in the spring 2024 issue of For All, the magazine of the Opportunity & Inclusive Growth Institute