Knowledge that Transforms

Financial advice can attenuate underinvestment but is costly, biased, and skewed towards the wealthy. AI-powered co-advisors could help deliver more scalable and affordable advice. To understand how, our vignette-based survey experiment compares the portfolio recommendations made by professional human advisors with GenAI large language models (LLMs) under biased and unbiased prompts. We document human financial advice projection whereby human advisors strongly project their own portfolios onto their clients. AI financial advice projection is prompt and model family dependent: ChatGPT is the least biased, while strong Gemini-Biased projection collapses when removing advisor demographics. LLMs are systematically more conservative than professional human advisors, recommending portfolios with lower Sharpe ratios that deliver up to 18% lower 20-year terminal wealth. However, human advisory fees erode much of this excess gain, with a 20-year breakeven fee of 1.03% p.a. Our results have direct implications for financial regulators, the advice profession, and LLM developers seeking to deploy AI-generated financial advice.

Drought can tighten agricultural credit conditions precisely when adaptation investments and access to working capital are most valuable. Using bank balance-sheet data merged with county-level U.S. Drought Monitor data for 2000–2020, we show that local credit markets exposed to drought experience significant declines in agricultural lending, with effects concentrated in severe episodes. These declines are strongest in markets served by geographically concentrated banks, especially single-county institutions, and weaker where lenders are more geographically diversified. In addition, we show that counties with greater irrigation intensity experience smaller lending declines during extreme droughts, while drought-related contractions are concentrated in counties with lower baseline crop resistance. Lending responses are also larger in counties with prior drought experience, consistent with persistent climate risk shaping local credit conditions. Our evidence highlights how climate risk, local adaptation, and bank structure jointly determine the availability of agricultural credit during drought episodes.

We investigate how trust shocks affect innovation networks through an incomplete contracting framework. Using academic misconduct cases in China (2015–2021) as an identification strategy, we construct a comprehensive dataset combining patent activities and venture capital investments. We document three key findings. First, academic misconduct triggers persistent declines in university–industry collaboration, reducing both joint patents and citations to university research. Second, affected firms strategically shift toward inter-firm R&D alliances. Third, trust shocks propagate to capital markets, with venture capitalists reducing investments in firms previously linked to universities involved in misconduct. Our findings highlight trust as an irreplaceable mechanism in innovation governance and demonstrate how trust breakdowns reconfigure contractual relationships and resource allocation in innovation networks.

Migrant welfare policies serve as crucial institutional levers for achieving sustainable urban development, yet the underlying mechanisms through which they influence firm value remain largely underexplored. This study proposes a new method for measuring the level of urban migrant welfare policies based on policy text scoring, and, using Chinese data, explores the impact of urban migrant welfare on corporate value and its underlying mechanisms. We find relatively robust evidence consistent with urban migrant welfare policies spilling over onto local corporate value. Mechanism tests suggest that these policies enhance corporate value by improving human capital allocation, promoting corporate social responsibility, and fostering innovation. The positive effect is more pronounced for firms in the growth stage and those operating in less competitive industries. Furthermore, we show that local fiscal capacity acts as a binding constraint: the value-enhancing effect of migrant welfare policies depends on sufficient fiscal resources for implementation. Our findings provide a new methodological perspective on quantifying policy text and elucidate the micro-foundations of how public policy shapes corporate performance.

Following the 2023–2024 Tokyo Stock Exchange (TSE) capital-efficiency initiative, investors favored low price-to-book (PBR) firms, particularly those with high return on equity (ROE). These effects are concentrated among persistently undervalued firms, increase smoothly with the degree of undervaluation rather than jumping at the regulatory threshold, and are robust to endogenous treatment assignment. Although firms’ value-enhancing plans were already publicly available in corporate governance reports, the TSE’s consolidated compliance list triggered strong market reactions, consistent with a salient informational event. We find no short-run improvements in realized profitability or upward revisions in analyst earnings forecasts, but significant declines in illiquidity and increases in valuation ratios among low-PBR firms. We also find little evidence of a persistent “price-of-shame” mechanism or strong industry-level spillovers, although spillovers through non-industry links cannot be fully ruled out. Overall, the evidence suggests that the reform primarily triggered investor-side repricing of chronic undervaluation rather than immediate changes in firm behavior.

Small leases and loans are excellent contracts to study the impact of information costs on the contract choice. Using a specially constructed dataset, we can directly compare the costs of leasing to borrowing for small firms. With a 15% average yield in our overall sample, we show that leases average about 12.5% while loans average 24%. After matching paired-samples of true leases and loans and correcting for selection bias, the differential is smaller but remains significant, while true and non-true leases show very little difference in yields. In our unique time series analysis, the average lease yields are significantly related to proxies for macroeconomic risk and demand factors. In sum, we reject the hypothesis that leasing is a more costly form of financing than “equivalent” borrowing and there are reasonable economic factors, related to enhanced collateral rights, that reduce information costs and account for the pricing differentials.

When tight monetary policy curtails domestic supply of credit and raises domestic borrowing costs, some firms can mitigate higher local borrowing costs by tapping global bond markets. This paper investigates whether this prediction holds for non-financial companies in the euro area. I first show that euro area firms exploit borrowing cost differentials between USD and EUR by issuing corporate bonds in USD when swap-adjusted U.S. dollar funding costs fall below euro rates. Using proxies for such opportunistic borrowing behavior, I then find that firms capable of seizing these opportunities in global corporate bond markets do not reduce their fixed capital investment to the same extent as other firms in response to monetary tightening. Further findings reveal that this differential investment response is not explained by differences in financial constraints or investment opportunities; instead, it reflects the ability to switch to lower cost offshore bond finance. Overall, the results underscore heterogeneity in the real effects of monetary policy and suggest that capital market openness can attenuate the domestic investment channel when global conditions allow lower cost funding abroad.

On average, wages in the finance industry are higher compared to the rest of the economy. Two explanations suggested for this finance wage premium are (1) the positive correlation between risk-taking and wages, and (2) industry differences in information technology intensity. Using a comprehensive worker-firm panel dataset for the Netherlands, we estimate wage models with additive worker and firm fixed effects, and compute the finance wage premium as the average of the firm fixed effects in an industry. We then relate the estimated cross-section of firm fixed effects to a range of firm characteristics, and find that information technology investment, the average level of educational attainment at a firm, and the complementarity of the two are the main drivers of the finance wage premium, while firm risk only makes a small contribution.

We study incentive contracts used by asset management firms in Norway, focusing on how bonus structures impact performance. The incentive contracts in our sample are heterogeneous, with firms rewarding fund managers based on both quantitative and qualitative targets. We find that higher potential bonuses tied to quantitative metrics, such as the information ratio, lead to better risk-adjusted performance at year-end. Managers at risk of missing bonus thresholds attempt to boost performance through portfolio adjustments, but these efforts backfire, resulting in worse outcomes in the latter part of the year.

Payment timeliness is crucial for the efficient functioning of supply chains. This paper examines the impact of extreme weather events on the timeliness of business establishments' payments to their suppliers, with a particular focus on Hurricane Katrina. We find that Katrina caused significant payment delays among affected establishments, with the impact being more severe for establishments whose parent firms were more exposed to Katrina, were financially constrained, or operated in markets with higher supplier competition. Following Katrina, these payment delays were associated with deteriorations in establishments' employment, survival, and exit outcomes, as well as in parent firms' supply-chain relationships and suppliers' financial performance. Our findings demonstrate that physical climate risks can profoundly affect the financial and real outcomes of customer and supplier firms through payment delays.