« Back to Results

Big Data / Machine Learning

Paper Session

Saturday, Jan. 4, 2025 2:30 PM - 4:30 PM (PST)

San Francisco Marriott Marquis, Yerba Buena Salon 8
Hosted By: American Finance Association
  • Anastassia Fedyk, University of California-Berkeley

Data Innovation Complementarity and Firm Growth

Orlando Gomes
,
Lisbon Accounting and Business School
Roxana Mihet
,
University of Lausanne and Swiss Finance Institute
Kumar Rishabh
,
University of Lausanne

Abstract

This paper examines how complementarity between a firm’s general innovation and data-security innovation affects firm outcomes in the modern data economy. Our theoretical model shows that when the importance of data and its protection increases, firms with high complementarity between data-security and non-data-security innovation enter a virtuous cycle. They take advantage of this complementarity to improve their broader predictive capabilities and extend their productivity frontier. Empirically, we propose a novel firm-level measure of data innovation complementarity based on the intersection of patent inventors who work on both data-security-related and non-data-security related patents. Leveraging the staggered introduction of Data Breach Notification Laws (DBNLs) across U.S. states as a quasi-exogenous shock, we provide robust empirical evidence that heightened incentives to protect in-house generated data activates this feedback loop. We find that firms with complementary data innovation processes
experience significant increases in (overall) innovation and profitability, by not only enhancing their in-house data security measures but also integrating these innovations across other domains. In contrast, firms without complementary data experience negative effects from the data protection laws. Our results highlight the dual role of data in driving firm-level market power and innovation dynamics.

Data Regulation in Credit Markets

Uday Rajan
,
University of Michigan
Yan Xiong
,
Hong Kong University of Science and Technology

Abstract

We study a credit market in which the lender makes decisions based on a borrower’s digital profile, and the borrower can manipulate the digital profile. We show that when the extent of data collected by the lender is observable, borrower manipulation increases when the lender utilizes more data in its underwriting models. Such manipulation worsens both the quality of the lender’s data and its lending decisions. Therefore, even if obtaining and analyzing additional data is costless, the lender will endogenously limit its own data coverage. Disclosure regulations can play a valuable role in allowing the lender to credibly commit to limiting its data coverage, and privacy regulations can benefit all borrowers, including those who choose to share their data with the lender.

What Are the Costs of Delayed Technology Adoption? Evidence from Italy’s Ban of ChatGPT

Jeremy Bertomeu
,
Washington University-St. Louis
Yupeng Lin
,
National University of Singapore
Yibin Liu
,
National University of Singapore
Zhenghui Ni
,
Renmin University of China

Abstract

On March 31, 2023, the Italian Data Protection Authority found ChatGPT in violation of privacy laws and banned the service in Italy, providing a natural experiment to evaluate the economic losses stemming from delayed adoption of generative AI. Italian firms with higher exposure to the technology suffer a negative market reaction of about 9% during the ban period compared to firms with lower exposure. The negative impact is greater for smaller and more newly established businesses. We further utilize job posting data to demonstrate that Italian firms in industries with high AI exposure experience a significant decline in full-time job postings, particularly for non-ChatGPT-replaceable roles. This decline persists into the post-ban period without signs of reversal, consistent with the path-dependency theory of resource accumulation. Moreover, firms in these industries also exhibit a decline in patent applications during and after the ban. These changes in real activities help explain the substantial value losses associated with even a short-term delay in technology adoption. Collectively, our results suggest the far-reaching impacts of delays in adopting revolutionary AI technologies.

Discussant(s)
Alex He
,
University of Maryland
Jian Li
,
Columbia University
Gregor Schubert
,
University of California-Los Angeles
JEL Classifications
  • G0 - General