Working Papers:

[1] Crafting an AI Compass: The Influence of Global AI Standards on Firms (2023)
with Zhe Wang

The adoption and commercialization of new technologies rely heavily on universally accepted principles, known as standards. Surprisingly, the role of standards in shaping business outcomes is largely unexplored. This paper provides a first examination of how  standardization of artificial intelligence (AI) influences business outcomes. By leveraging hand-collected data on global AI standards, we provide insights into their creation, content, and shed light on the countries leading or lagging in global AI standardization efforts. Our findings reveal that establishing universal rules and guidelines for AI has a first order impact on corporate investment. Standardization of machine learning methods, programming languages, protocols for big data,  guidelines for data interchange, and interoperability of industrial data fuel capital and R&D investments. Conversely, frameworks for ethical, societal, and privacy considerations in AI have distortionary effects. Overall, AI standards have a positive 'ripple effect' on firm value, gradually amplifying as the standards mature and their impact on firms deepens. 

[2] Financial Consequences of the Belt and Road Initiative (2023)

As the largest ever infrastructure project, China’s Belt and Road Initiative (BRI) is expected to reshape the global economy for the coming decades. This paper provides the first analysis of BRI's effects on financial markets and real economic activity in Europe. It exploits the opening of a subway tunnel under Istanbul's Bosporus Strait that geographically positions nearby countries on BRI's economic corridor. Governments of impacted countries sharply increase sovereign debt issuance and devote more resources to collective consumption spending. Meanwhile, their strategic infrastructure investments stagnate, news-based macroeconomic sentiments for inflation, financial stability  and economic uncertainty worsen, and sovereign yields rise. Businesses in these countries issue less debt, lower capital investments, and observe reductions in their valuations.  Overall, my findings highlight the distortionary effects of BRI on European businesses  and raise concerns about the effectiveness of BRI-driven government spending in Europe. I provide additional findings on developmental aid, international trade, and BRI program membership that highlight China's growing influence over corridor economies.

Presentations: Fifth Conference on Law and Macroeconomics, 2022 New Zealand Finance Meeting,  CMU-PSU-Pitt Finance Conference, MFA 2023 Annual Meeting, 2023 FMA Annual Meeting.

[3] An Anatomy of Cryptocurrency Sentiment (2023)
with Charles Cao, Giang Nguyen, and Qiang Wang

In this paper, we investigate the impact of market sentiment on cryptocurrency returns. To accomplish this, we use a novel dataset that captures a multitude of attitudes, moods, and emotions extracted from a vast amount of news and social media content. Our findings indicate that social media sentiment significantly predicts crypto returns, while sentiment from news media does not. Additionally, we observe that fundamental events play a role in shaping sentiment, but market exuberance---sentiment unrelated to fundamental events---is a strong and robust predictor of returns. Furthermore, we find that market exuberance is positively related to momentum return but does not positively predict volatility. This suggests that sentiment influences returns through price perception and demand shocks rather than the risk premium channel. Overall, our research highlights the importance of sentiment in understanding and predicting cryptocurrency market dynamics.

Presentations:  Penn State University*, CMU-PSU-Pitt Finance Conference*, 2023 Early Career Women in Finance Conference*, 2023 FMA Annual Meeting*.

* denotes presentation by co-author

Media: Les Affaires.

[4] Macro sentiment and hedge fund returns (2023)
with Mustafa Caglayan, Tim Simin, and Le Zhao

Using a novel dataset, we show media sentiment on growth, inflation, unemployment, and sociopolitical conditions predict hedge fund returns. We blend these into a macro sentiment index (MSI) and show that funds betting against sentiment generate higher risk-adjusted returns. This result is robust to orthogonalizing MSI against popular sentiment indices. We demonstrate MSI as a state variable in Intertemporal Capital Asset Pricing Model (ICAPM). Our sentiment ICAPM produces 22% fewer significant alphas relative to the standard nine-factor model. Overall, our results highlight differences between media-based and popular outcome-based sentiment measures and suggest a role for sentiment in asset pricing models.

Presentations: 2022 FMA Annual Meeting in Atlanta*.

* denotes presentation by co-author

[5]  Judicial Ideology and Business Dynamics (2023)
with Matthew T. Gustafson 

Using staggered changes in U.S. circuit court ideology and counties along circuit borders, we examine the effect of judicial ideology on the business start-up decision. We find evidence that shifts toward more liberal courts, which are commonly viewed as less pro-business, are followed by immediate reductions in the number of firm and business establishments in an area. This reduction is largest in the year after the ideology change and driven primarily by fewer entries from de novo firms. An industry decomposition reveals that retail firms, which are traditionally viewed as litigation sensitive, are the primarily drivers of our results. Overall, our study shows that start-ups weigh an area's judicial risk when deciding to open a business.  

Presentations: Iowa State University*, Third Annual Conference on Law and Macroeconomics, University of Miami*,  MFA 2021 Annual Meeting, University of Bath*, Vrije Universiteit Amsterdam, 2021 American Law and Economics Association (ALEA) Annual Meeting in Chicago, 2023 FMA Annual Meeting. 

* denotes presentation by co-author

[6] Privacy Laws and Value of Personal Data (2022)
with Ilja Kantorovitch and Roxana Mihet

We analyze how the adoption of the California Consumer Privacy Act (CCPA), which limits consumer personal data acquisition, processing, and trade, affects voice-AI firms. To derive theoretical predictions, we use a general equilibrium model where firms produce intermediate goods using labor and data in the form of intangible capital, which can be traded subject to a cost representing regulatory and technical challenges. Firms differ in their ability to collect data internally, driven by the size of their customer base and reliance on data. When the introduction of the CCPA increases the cost of trading data, sophisticated firms with small customer bases are hit the hardest. Such firms have a low ability to collect in-house data and high reliance on data and cannot adequately substitute the previously externally purchased data. We utilize novel and hand-collected data on voice-AI firms to provide empirical support for our theoretical predictions. We empirically show that sophisticated firms with voice-AI products experience lower returns on assets than their industry peers after the introduction of the CCPA, and firms with weak customer bases experience the strongest distortionary effects. 

Presentations: University of Lausanne*, ESSEC*, 2022 Next Generation of Antitrust, Data Privacy and Data Protection Scholars Conference (NYU), 15th Digital Economics Conference at Toulouse School of Economics*, NBER Economics of Privacy Conference, AEA 2023.

Slides: NBER Slides can be found here.

* denotes presentation by co-author

[7] Capitol punishment: Effects of congressional seniority ranks on corporate outcomes in the U.S. (2021)     

I study the effects of congressional seniority ranks on corporate outcomes in the U.S. I show that firms located in districts of lower-ranked legislators exhibit weaker sales to government, corporate downsizing, and lower valuations. These firms are also more exposed to civil disorder, including attacks that specifically target government offices, infrastructure, and employees. Seniority lotteries in House committees suggest that the links between congressional seniority ranks and corporate outcomes are causal.    

Presentations: University of Amsterdam, Analysis Group, Cornerstone Research, Einaudi Institute, George Washington University, Said Business School at University of Oxford, Smeal College of Business at Pennsylvania State University, Simon Business School at University of Rochester, Sabanci University, Stanford University GSB.

[8] Can Enforcements Stop Illicit Sellers on E-Commerce Platforms? (2020) 

This paper provides the first exploration of how illicit sellers operate on e-commerce platforms and how they respond to enforcements. I use a novel data set of 71 million illicit — i.e., fraudulent, counterfeit, or replica — items that were removed from online marketplaces. By using natural language processing and computer vision techniques on these products and by tracking business activities of the illicit sellers, I identify a large number of similar but previously unnoticed illicit products (UIPs) that are currently sold online. For each illicit product that was previously removed, I detect 16.91 UIPs. Of these, 84% remained on the market during the one-year period after the removal of the initial illicit product. Nonetheless, the total market value of these products decreased by up to 80% after enforcements. My findings suggest that enforcements against illicit products on e-commerce platforms encourage separating equilibria, in which illicit sellers have weaker incentives to pool with authentic producers than to be revealed as low-quality producers.

Papers at the R&R Stage:

[1] Choose Your Battles Wisely: The Consequences of Protesting Government Procurement Contracts (2022)
with Jess Cornaggia and Kimberly Cornaggia

We examine the relation between a firm’s successful protest of a government agency’s conduct or terms of a procurement contract and the amount of business the firm conducts with the government going forward. We find firms receive fewer and less valuable government contracts, face more contract cancellations, and experience significant reductions in sales growth and employee growth. Despite widespread belief, successful bid protesters do not delay government procurement due to lengthy dispute resolutions. Overall, we provide the first analysis of corporate interactions with the United States government bid-protest system. 

Presentations: 2020 American Finance Association Meetings (San Diego),  MFA 2020 Annual Meeting in Chicago, 2020 American Law and Economics Association (ALEA) Annual Meeting in Chicago, Clemson University*,  Emory University*, Pennsylvania State University (Smeal College of Business and Harrisburg),  Emory University*, and Temple University (Fox School of Business)*.

* denotes presentation by co-author