How do we overcome the lack of data? By combining causal inference with new confidential data sources to study several different types of illicit financial flows and the policies aimed at stopping them. Many of these data sources come from various forms of leaks: whistleblowers or hackers who have turned the data over to journalists or transparency groups for further publication. This data has a significant advantage over publicly available data: because it covers behavior by persons who expected to remain hidden, it allows for the precise estimation of many outcomes and relationships that a researcher cannot normally observe.
As such data is not without its own unique challenges and methodological considerations, I have dubbed this approach “investigative economics,” the intersection where public economics, forensic economics and investigative reporting intersect. By directly processing the unique, confidential data with the end goal of estimating an economic outcome of interest (such as the stock of foreign-held real estate) or a causal parameter of interest (e.g. the average treatment effect of ownership transparency on the flow of foreign-held real estate), our approach avoids the above pitfalls.
The name arises because we must combine multiple sources of data, both within the main confidential source (e.g. processing beneficial ownership information and connecting it to different firms) and in liking further datasets (e.g. publicly provided or government-held data) as in investigative reporting. Investigative economics can be thought of as a subset of forensic economics, a field focused on using a variety of methods to uncover “hidden” behavior. The goal is to marry the statistical rigor applied econ with the hand-crafted case approach of investigative journalism.
In 2024, I along with my colleagues at NMBU were awarded a Norwegian Research Council Project for Experienced Scientists (FRIPRO) grant for NOK 10,302,000 (roughly $1m) to work on this topic.