Abstract:

​

 We define FinTech as banks’ integration of technology into the broker intermediation model. Unlike traditional dealer banks relying on leveraged balance sheets for intermediation, FinTech banks adopt a technology-driven broker model. Using nonlinear and machine learning algorithms, we develop an empirical “FinTech score” that shows on-balance sheet lending for low-FinTech-score banks versus securitization, brokered deposits, and non-interest income for high-score banks. Using two complementary measures of operational efficiency, we find that this technology-driven shift in business models (either holistically or via mergers) helps explain reductions in the cost of financial intermediation. Our bank-specific, time-varying FinTech score provides insights into the distribution of FinTech integration into U.S. banking. Link to paper

Abstract:

​

Organization capital (OC) constitutes a firm’s unique, hard-to-mimic, hard-to-transfer culture, internal knowledge and language, policies and procedures, growth opportunities and information technology, brand name and any other aspects that are not directly related to the production process. In order to disentangle value enhancement from agency costs inherent in the OC, we calibrate systematic risk premiums from the loss of managerial OC rents by decomposing OC into: (1) portable economic rents in the form of disclosed executive compensation and (2) a strategic component comprised of undisclosed, non-portable OC rents mixed with agency costs. Shareholders in firms with high strategic component (not the portable component) earn an annual systematic risk premium of 4.6%. Disclosures revealing the breakdown between OC and agency costs in the strategic component (earnings surprises, managerial obfuscation and stock mispricing) as well as corporate governance eliminate this risk premium by revealing the OC rents in strategic component and increasing OC portability. Link to paper

Abstract:

​

Does relationship bank oversight reduce firm default risk and improve firm operational efficiency? I find that a new loan from a relationship bank reduces the default probability and increases the efficiency of a borrowing firm, benefiting both banks and borrowers. Moreover, inefficient and less creditworthy firms experience the highest reductions in their default risks and improvements in their efficiencies in the years following new relationship bank loans. Further, these benefits are disrupted when the relationship bank is acquired. Link to paper

Abstract:

​

Generative Artificial Intelligence (GenAI) has the potential to transform many industries and is exposing workers in a variety of organizational roles to uncertainty about the future of their position. Using data from Glassdoor reviews, we investigate how the introduction and diffusion of this technology influences worker satisfaction and find evidence of a significant decrease in satisfaction. We find that this decline is concentrated in firms with high GenAI exposure following the release of ChatGPT. This effect is driven by increased job uncertainty and anxiety, evidenced by text analysis of employee reviews, rather than by immediate changes in productivity. The negative impact is significantly more pronounced in smaller, less profitable firms and, paradoxically, within the technology sector.

Abstract:

​

This paper examines how the Paris Agreement influenced banks’ financed emissions through their syndicated loan portfolios. We focus on two key channels—divestment from high-emitting borrowers and engagement through screening and monitoring—while differentiating between relationship (existing) and non-relationship (new) lending, sectoral exposures, and carbon pricing regimes. We find strong evidence of divestment but limited engagement. Financed emissions from new energy-sector borrowers fell by over 20%, driven by both reduced lending volumes and a shift toward lower-emitting firms. In contrast, emissions in relationship lending portfolios remained largely unchanged, indicating banks’ reluctance to sever ties with existing, profitable borrowers.

Banks’ responses varied by borrower regulatory environments. All banks curtailed financing to non-relationship borrowers in regulated countries, while Brown-regulated banks also divested from relationship borrowers in unregulated countries. Engagement was evident mainly in energy-sector relationship lending to borrowers located in countries with carbon pricing regimes. Borrowers’ broader financing environments also influenced outcomes. Brown banks reduced financed emissions in energy-sector relationship portfolios only when borrowers lacked ESG-oriented investors. In contrast, when borrowers had ESG stakeholders, relationship lending sometimes saw increased emissions. Overall, the Paris Agreement induced selective portfolio decarbonization,

with divestment concentrated in new, high-emitting borrowers rather than through deeper engagement with existing clients. These findings highlight the limits of voluntary bank-led engagement and reveals that banks’ climate strategies depend critically on lending relationships, borrower characteristics, and the surrounding regulatory and investor contexts.