Research

Who am I?

I’m a PhD student in Economics at Lund University, Sweden. I successfully defended my thesis in June 2019. Currently working as a Data Scientist at Danmarks Nationalbank.
My research interests are machine learning, network effects, household finance and economics of education. Download my CV.

.

Papers I’m working on

Bouncing with the Joneses? A Neural Network Approach to Consumption Neighborhood Effects
If one of your neighbors decides to purchase a good, does that affect your decision to buy the same good? I study this question using household level data collected in a novel way. I use an image classfication algorithm to process a large set of aerial photos in order to infer household ownership of a visible durable good, specifically a trampoline. To estimate the neighborhood effect, I use a neighborhood fixed effects approach together with exogenous variation stemming from new neighbors moving in. I find that neighborhood effects are present, but only over short distances (the five closest neighbors) and only for prior non-owners. Information transmission between neighbors seems to be an important underlying channel. A back-of-the-envelope simulation suggests that the effect is large enough to have a significant impact on demand for this particular product.
[Paper] [Code]

.

Cashing out? Children and Financial Risk. Joint with Sara Mikkelsen.
We document how the financial portfolios of parents change in response to the birth of a child. To identify dynamic effects around child birth, we use an event study approach with a matched treatment and control sample and a novel implementation of the same sex instrument. We find that parents reduce their financial risk by a reduced propensity to participate in risky financial markets.
[Paper]

.

High-Speed Broadband and Academic Achievement in Teenagers:
Evidence from Sweden.
Joint with Martin Nordin.
We examine the effects of super-fast internet connections at home on the grades of students in upper secondary school. We link register data on around 250,000 students to local levels of access to optic fiber broadband and estimate the effect of broadband on student GPA. We show that going from zero to full coverage in the student’s parish of residence causes a GPA reduction of 3 to 6 percent.
[Paper]

.

Publications

Estimating the Impact of Agrienvironmental Schemes on Nutrient Leaching.
Joint with Martin Nordin. 2018. Land Use Policy, vol 74, pp. 388-398.
We estimate the effect of Agri-Environmental Schemes (AES) on nutrient runoff using abatement data and water samples on a large scale. We identify all farms located upstream from a given water sampling site. Using a watershed fixed-effect model we estimate the effect of within-watershed variation in AES payments on nitrogen and phosphorus concentrations in downstream water samples. Higher uptake of a wetland subsidy is associated with reduced P and N-concentrations.
[Paper]