We use electronically-filed federal tax records from tax year 2019 to document the first
available evidence on the short-run response of financial capital to the Opportunity Zone (OZ)
program, a federal place-based policy that provides tax incentives for capital investments in
more than 8,000 low-income neighborhoods across the United States. We observe $18.9 billion
of aggregate OZ investments from electronic filers in tax year 2019. While these data do not
yet incorporate paper filers, several suggestive patterns emerge. First, OZ capital appears
highly spatially concentrated: 84% of designated Opportunity Zone tracts in our sample receive
zero OZ investment, and among those that do receive investment, the distribution is highly
skewed. Second, we correlate OZ capital with tract-level demographic characteristics and
show that, although tracts that receive investment are disadvantaged relative to the general
US population, they appear well-off relative to other OZ tracts that do not receive investment.
Among OZ tracts, investors report equity and property investments in neighborhoods with
relatively higher incomes, home values, educational attainment, and pre-existing income
and population growth. These tracts have also experienced significant changes in their
demographic composition over the past decade. Finally, we geocode the universe of individual
and business tax records to construct novel tract-level measures of wages, employment,
commuting, firm growth, and real investment that we plan to use to evaluate the causal effect
of the OZ tax subsidies as more comprehensive data become available. We demonstrate that
these estimates closely match corresponding measures from publicly available Census data.
We explicitly discuss limitations of the data, and will update this working paper as more
comprehensive data become available.

Kennedy, Patrick and Wheeler, Harrison