Entrepreneurship requires energy and creativity as well as business acumen. Some factors that contribute to entrepreneurship decline with age, but business skills increase with experience in high-level positions. Having too many older workers in society slows entrepreneurship. When older workers occupy key positions, they block younger workers from acquiring skills. A theory is formulated and tested using the Global Entrepreneurship Monitor data. A one standard deviation decrease in a country’s median age increases new business formation by 2.5 percentage points, which is about 40 percent of the mean rate. Furthermore, older societies have lower rates of entrepreneurship at every age.

We develop FinBERT, a state‐of‐the‐art large language model that adapts to the finance domain. We show that FinBERT incorporates finance knowledge and can better summarize contextual information in financial texts. Using a sample of researcher‐labeled sentences from analyst reports, we document that FinBERT substantially outperforms the Loughran and McDonald dictionary and other machine learning algorithms, including naïve Bayes, support vector machine, random forest, convolutional neural network, and long short‐term memory, in sentiment classification. Our results show that FinBERT excels in identifying the positive or negative sentiment of sentences that other algorithms mislabel as neutral, likely because it uses contextual information in financial text. We find that FinBERT's advantage over other algorithms, and Google's original bidirectional encoder representations from transformers model, is especially salient when the training sample size is small and in texts containing financial words not frequently used in general texts. FinBERT also outperforms other models in identifying discussions related to environment, social, and governance issues. Last, we show that other approaches underestimate the textual informativeness of earnings conference calls by at least 18% compared to FinBERT. Our results have implications for academic researchers, investment professionals, and financial market regulators.

Geographic dispersion of depositors, borrowers, and banks may prevent funding from flowing to high-loan-demand areas, limiting credit access. Using bank-county-year-level data, we provide evidence of geographic imbalance of deposits and loans and develop a methodology for investigating the contribution to this imbalance of branch networks, market power, and scope economies. Results are based on a novel measure of imbalance and estimation of a structural model of bank competition that admits interconnections across locations and between deposit and loan markets. Counterfactual experiments show branch networks, scope economies, and local competition affect credit flow to disadvantaged markets.

We examine the significance and uniqueness of individual-pair relationships cultivated through repeated loan interactions. Using a hand-collected dataset compiled of borrowing manager and loan officer information, we find that individual-pair relationship loans are associated with a cost-of-debt reduction of between seven to 13 basis points. We also document that the relationship has an economic impact even when other affiliations, for example, institutional pairs, social ties, cultural proximity, and gender, are considered. Individual-pair relationships matter because they furnish lenders with useful soft information, especially when the firm has a poor hard information environment or when the bank and loan officer rely less on hard information. In addition, we find that individual-pair relationship loans have fewer rating downgrades, suggesting that accumulated soft information leads to better loan quality. Collectively, our results highlight the unique value of sustained professional engagement between two individuals in the lending process.

In common-law systems, firms’ litigation risk depends both on written laws and how courts interpret these laws. Using 321 US circuit court rulings, we introduce a novel measure capturing courts’ attitudes toward defendants in securities lawsuits. Our results confirm that financial misreporting firms in more defendant-friendly circuits face fewer lawsuits. Consistent with lower expected litigation costs, firms in these circuits face less negative market reactions when misreporting is revealed, invest less in preventing misreporting, and are more likely to engage in aggressive misreporting. We conclude that defendant-friendly precedents reduce firms’ legal liability and worsen their financial reporting quality.

Unlike labor income, human capital is inseparable from individuals and does not completely accrue to creditors. Therefore, human capital investment is more resilient to “debt overhang” than labor supply. We develop a dynamic model displaying this difference. We find that while both labor supply and human capital investment are hump-shaped in household indebtedness, human capital investment declines less aggressively as indebtedness builds up. Importantly, because human capital is only valuable when households expect to supply labor, the greater reduction in labor supply due to debt overhang back-propagates into ex-ante human capital investment. We provide empirical support for the model.