I am an innovation and industrial policy expert and PhD Candidate in Economics at Humboldt University in Berlin (expected: 2026).
My policy work has focused on industrial policy in high-tech sectors (chips, AI) and innovation policy (tech transfer, how to organize innovation funding).
My research interests include the economics of innovation, science of science, and digitalization. I primarily use methods from experimental and behavioral economics in my research.
You can download my CV here. (Academic CV here)
“The Elusive Returns to AI Skills: Evidence from a field experiment", with Anastasia Danilov. [Under review]
As firms increasingly adopt Artificial Intelligence (AI) technologies, it remains unclear how they adjust hiring practices for skilled workers This paper investigates whether AI-related skills are rewarded in recruitment by conducting a large-scale correspondence study in the United Kingdom. We submit 1,185 résumés to vacancies across a range of occupations, randomly assigning the presence or absence of advanced AI-related qualifications. These AI qualifications are added to résumés as voluntary signals and are not explicitly requested in the job postings. We find no statistically significant effect of listing AI qualifications in résumés on interview callback rates. However, a heterogeneity analysis reveals some positive and significant effects for positions in Engineering and Marketing. These results are robust to controlling for the total number of skills listed in job ads, the degree of match between résumés and job descriptions, and the level of expertise required. In an exploratory analysis, we find stronger employer responses to AI-related skills in industries with lower exposure to AI technologies. These findings suggest that the labor market valuation of AI-related qualifications is context-dependent and shaped by sectoral innovation dynamics.
“Improving the Selection of High Social-impact Technology Firms with AI: Evidence from a Field Experiment", with Anastasia Danilov, Albert Banal Estañol, and Ina Ganguli
AI tools are increasingly being used in the evaluation of research and innovation funding. This trend raises important questions for public agencies funding innovation, both related to the overall impact of AI tools and which designs are most effective. We investigate how human evaluators incorporate algorithmic advice related to a special domain of expertise: the potential social impact of high-tech projects seeking R&D funding. We run a randomized experiment in which professional and volunteer assessors evaluate project proposals and are exposed to different forms of AI-generated feedback. The design allows us to distinguish between the extensive margin of updating (whether evaluators revise their scores) and the intensive margin (by how much they do so). Using detailed proposal-level and evaluator-level data, we analyze how advice presentation affects evaluators’ responsiveness to algorithmic input.
“Principals' Expertise and Quality of Bids in Procurement Auctions", with Anastasia Danilov
“Green Innovation in Small Firms: Evidence from a field experiment", with Anastasia Danilov
This paper investigates the effectiveness of a €12,000 voucher scheme for small firms in Austria to undertake decarbonization activities. We exploit the fact that access to the voucher was randomized due to oversubscription to study the causal impact of the program. A survey of applicants reveals that funded firms are significantly more likely to carry out the project they applied with and to report plans to further decarbonize. Moreover, we study whether the introduction of randomization in 2023 changed the type of applicants or the kind of proposal submitted. Using data from public registries as well as the implementing agency, we find no statistically significant differences after the introduction of the lottery. The only exception is that applicants under randomization are more likely to be first-time applicants and be based in the capital.
“Promoting internalization of firms through E-Commerce: Evidence from a field experiment in Tunisia", with Florian Münch, Fabian Scheifele, and Amira Bouziri
"Attracting Firms to Government Programs: Theory and Evidence from Randomized Controlled Trials in Tunisia", with Florian Münch, Fabian Scheifele, Amira Bouziri, and Kaïs Jomaa. IGL Working Paper No. 24/01. [Link]
Governments spend over a billion US dollars annually on firm support programs, yet application rates are low and outcomes modest. Attracting enough and the right firms may alter the program’s effect and statistical power to detect it. Yet, we document that most firm program evaluations don’t report recruitment strategies. We conduct two email experiments involving 5000 SMEs while recruiting for two export support programs in Tunisia, tracking each communication channel’s contribution to registrations. In experiment 1, we find goal-specific messages targeting firms’ supply or demand side constraints attract fewer but better-performing firms. In experiment 2, we find an influencer video emphasizing program benefits attracts better-performing female-led firms, while reducing participation costs via free childcare attracts less-performing firms managed by younger female entrepreneurs with children. Finally, we show open communication channels attract more underrepresented firms. In general, the findings suggest recruitment strategies substantially impact sample size and composition.