CELINE FEI
  • Research
  • More
Celine Yue FEI
Picture
Welcome!

Department of Economics,
University of Mannheim

Contact Information:
L 7, 3–5 – Room 341 (3rd floor)
68161 Mannheim
​Email: [email protected]

I am teaching the Financial Economics (undergrad) and Fintech (master) courses at Mannheim.

My Master Course: Fintech and AI in Finance is part of the online initiative of the ENGAGE.EU alliance (LUISS University, NHH Norwegian School of Economics, Tilburg University, Université Toulouse 1 Capitole, University of National and World Economy, WU Vienna, University of Mannheim). 
​​
Research Areas
  • Venture Capital, ​Private Equity
  • ​Fintech and AI in Finance
  • Diversity, Equity, and Inclusion (DEI)
  • Matching Models

I worked in the Finance Area at Kenan-Flagler Business School of UNC at Chapel Hill before. I obtained my Ph.D. in Economics from Toulouse School of Economics. 
CV

​
Research Papers
Directed Financing of Innovation: From IPO Reforms to Venture Capital, with Ulrich Hege, Xiao Jia
  • Selected presentations: 2025 AFFECT, 2025 Entrepreneurial Finance (ENTFIN, scheduled, *), 2025 Aarhus Workshop on Strategic Interaction in Corporate Finance (scheduled), 2025 MaCCI IO Day
Abstract: We study the impact of new entrepreneurial IPO markets on the technology direction and frequency of venture capital (VC) financing rounds. We exploit the regulatory differentiation between favored and opposed business activities for new IPO listings in China's two most important market introductions, ChiNext in 2009 and STAR in 2019, and use deep learning methods to build a treatment intensity measure of textual closeness between mandatory business descriptions of startups and granular business activities in IPO rules. Using a difference-in-differences design, we find that after the launch of ChiNext and STAR, VCs invest more in companies with business activities favored by the listing rules, and less in companies with business activities opposed by them. In addition, VCs change their investment policies by expanding funding rounds for companies with performance credentials in the form of patents, trademarks, and certificates, but at the same time prefer companies with a less established record.
​
​* Presented by coauthor ​
Portfolio Management in Private Equity, with Gregory W. Brown and David T. Robinson R&R at Journal of Financial Economics (JFE)
NBER Working Paper No. w31664
  • Selected Presentations: 2023 AFA, 2023 FIRS, 2023 Columbia Private Equity Conference (*), 2023 Mitsui Finance Symposium (Michigan Ross), 2023 Summer Institute of Finance Conference (Shanghai Jiao Tong SAIF), 2023 Lapland Investment Fund Summit​ (Aalto)​, 2023 SKEMA Corporate Restructuring Conference, ​2022 Finance, Organizations and Markets Conference (USC Marshall), 2022 WashU St. Louis Annual Conference on Corporate Finance (*), 2022 UNC Annual Private Equity Research Consortium ​(*), 2022 Conference on Financial Economics and Accounting​​ (GSU), 2022 Society of Quantitative Analysts conference (*), 2021 IPC Research Symposium (*)
  • ​Matlab Code
Abstract: Private equity investments are typically made through closed-end fund structures that tie performance-based compensation together across investments. This creates incentives for portfolio formation considerations to affect investment decisions. With novel, deal-level data covering a large sample of global investments, we show that these portfolio considerations are important for understanding investment behavior and fund returns. The largest investments in PE funds typically have the lowest returns on average but are also the least risky. Returns and risk are both increasing in industry concentration. Later investments are safer when they follow early successes, suggesting that GPs act on incentives to ''lock in'' early performance-based compensation. These findings show that GPs use portfolio construction, not just deal selection, to seek risk-adjusted fund-level returns.
​* Presented by coauthor​
What Drives Racial Minorities to Use Fintech Lending? R&R  Journal of Financial and Quantitative Analysis (JFQA)​
  • Selected Presentations: ​2021 NBER Entrepreneurship Working Group, 2023 MFA, 2023 Cambridge Centre for Alternative Finance Annual Conference, 2023 OFR Rising Scholar Conference, 2023 University of Oregon Summer Finance Conference​
​ Abstract: Using a national sample of Paycheck Protection Program restaurants, I find evidence of a more negative minority-non-minority rating gap for fintech lenders, which suggests taste-based discrimination of Becker (1957). To quantify various channels' magnitude, I estimate an empirical matching model where I find that fintech-minority matches generate more value than other matches. Moreover, this fintech-minority additional value channel is the most important channel in explaining racial disparities in fintech usage. Disabling this channel reduces minority borrowers' usage of fintech by approximately 70%. Disabling lending relationships and bank branch channels only reduces minority borrowers' usage of fintech by less than 2%.
​​
Non-Peer Reviewed Publications

Chapter on Entrepreneurial Finance
, ​with Xiaoyun Yu
Sage Handbook of China Economy and Financial System​ (To be Published in 2026)
​
​

Work-in-Progress

Investor attention and fund-flow in venture capital : A Large-Language Model Analysis, with Allen Hu

Impact of AI in Financial Labor Market, with Tania Babina


Can Government Foster the Development of Venture Capital?
Best Paper Award at 2022 China International Forum on Finance and Policy (CIFFP)

Journal Referees
Management Science, ​Review of Financial Studies, ​Journal of Corporate Finance

​Teaching

Univ of Mannheim (Instructor): Financial Economics (undergrad); Fintech and AI in Finance (master)
UNC Kenan-Flagler (Instructor): Corporate Finance BUSI408 (undergrad); evaluation 4.6/5
Toulouse School of Economics (TA): Macroeconomics PhD Core, Macroeconomics, Econometrics
​Discussion [Link]

  • Research
  • More