« Back to Results
AFA PhD Student Poster Session
Friday, Jan. 7, 2022
7:00 AM - 6:00 PM (EST)
Saturday, Jan. 8, 2022
7:00 AM - 6:00 PM (EST)
Sunday, Jan. 9, 2022
7:00 AM - 3:00 PM (EST)
American Finance Association
A Machine Learning Based Anatomy of Firm Level Climate Risk Exposure
We construct firm-level climate risk exposures by utilizing two natural language processing
techniques (LDA and word2vec) on quarterly earnings conference call transcripts.
This unsupervised learning method automatically generate five topics, all aligned with
popular concerns about climate change. We then conduct empirical analysis on one of
the topics that put high weight on words about natural disaster. This topic has a significant
negative association with firm's sales growth and profitability, indicating that our
measure exactly capture firm level disaster exposure. Moreover, firm with higher disaster
measure tend to earn higher stock returns in the future years. Appropriate long-short
portfolios based on this topic generates positive return, which cannot be explained by
common risk factors and other firm characteristics.
A Macro-Finance model with Realistic Crisis Dynamics
What causes deep recessions and slow recovery? I revisit this question and develop a macro-finance model that quantitatively matches the salient empirical features of financial crises such as a large drop in the output, a high risk premium, reduced financial intermediation, and a long duration of economic distress. The model has leveraged intermediaries featuring stochastic productivity and a regime-dependent exit rate that governs the transition in and out of crises. A model without these two features suffers from a trade-off between the amplification and persistence of crises. My model resolves this tension and generates realistic crisis dynamics.
Global Common Ownership, Control and Market Power
I examine the simultaneous ownership of equity of companies by the same investor, i.e. common ownership, in 39,967 publicly listed corporations in 125 countries in the last two decades. Existing empirical assessments typically rely on common ownership by asset managers in selected industries in the U.S., obscuring different forms of common ownership between countries. I document the prevalence of common ownership by asset managers, business groups, and cross-ownership between companies, both within and across industries. Using the general equilibrium framework of Azar and Vives (2021), I measure intra-industry and inter-industry common ownership and empirically investigate their respective effects on market power. Preliminary results suggest that intra-industry common ownership increases firm markups whereas inter-industry common ownership decreases them. Managers are incentivised to behave anti-competitively when common ownership increases within the same industry, and pro-competitively when common ownership increases across industries.
Asset Pricing via Graph Neural Networks
Arbitrage Pricing Theory claims that there is an approximately linear relation between excess returns and risk factors in a diversified and non-arbitrage market. Despite an often clear linear structure, a major challenge is that true factors driving the changes in returns are unknown.Over the past half-century, researchers kept proposing methods, ranging from economic intuition to purely statistical techniques, and building a zoo of factors. However, how can we fully use data from various sources and discover effective factors? In order to solve the puzzle, we aim to develop a dynamic machine learning framework, incorporating macroeconomic factors, asset specific characteristics and cross impact among assets, to estimate risk factors and risk exposures in a linear multi-factor model. Following recent literature, we apply neural network models, which take characteristic and macroeconomic data as inputs, to learn low-dimensional embeddings as latent risk factors. In order to model cross impact, we construct networks whose vertices are individual assets. Taking advantage of research on correlation networks over the past two decades, we define cross impact as pairwise correlation between assets, calculated from rates of returns using a moving window approach. Then we apply information filtering graphs to build edges in the stock networks. Recent developments in graph neural networks provide us with tools to include complex relations among stocks in the framework. Here, we employ graph convolutional networks(GCN) to extract latent variables from collected data. We conducted empirical studies on US stocks from 1959 to 2016 by building a small network of 120 stocks which exists over the whole study period based using monthly returns. The model with GCN we propose achieved slightly better profits during backtesting over the test period than a standard conditional autoencoder. Therefore, including networks of stocks as proxies for cross impact can be beneficial for predicting asset returns.
Asymmetric Information, Patent Publication, and Inventor Human Capital Reallocation
How does patent publication affect inventor human capital reallocation? Leveraging (i) the American Inventor’s Protection Act (AIPA) that requires patent applications to be published within 18 months of filing rather than when granted, (ii) plausible quasi-random assignment of patent examiners, and (iii) a large dataset tracking inventors’ career paths, I find that the expedited patent publication helps inventors switch employers. Additional analyses suggest that increased mobility is driven by expanded outside employers’ information and reduced information asymmetry: the effect is more pronounced among inventors of high-quality, with scarcer existing information, or in technologies where patents are more informative. Suggestive evidence indicates a positive effect on enhancing inventor-firm matching and patenting output. This study highlights an important yet overlooked facet of patent publication: inventor human capital reallocation.
Bank Competition and Personal Bankruptcy: Evidence from Large Bank Mergers
This paper studies the role of credit market competition in explaining consumer bankruptcy filings. I exploit variation in bank competition induced by large bank mergers to establish that personal bankruptcy rates are significantly higher in more competitive local banking markets. I also find that higher competition prompts banks to take more risks by increasing credit supply and lowering their credit standards. Finally, using bank balance sheet data, I demonstrate that banks that operate in more competitive counties have higher credit supply and exhibit a greater loan loss rate, consistent with the bank risk-taking channel.
Bank Credit Provision and Leverage Constraints: Evidence from the Supplementary Leverage Ratio
We causally identify the implications of relaxing the Supplementary Leverage Ratio in April 2020 for bank balance sheet composition and credit provision. Our findings suggest that this risk-invariant leverage ratio was binding for banks, weakly affected bank liquidity provision in Treasury markets, and strongly affected banks' portfolio composition across asset classes, amounting to a shift of banks' loan supply schedules. The increase in lending is driven primarily by real estate and personal loans, and to a lesser extent by commercial and industrial loans. We additionally provide evidence that the relaxation allowed banks to increase their repo borrowing from cash providers. Our evidence highlights that countercyclical relaxation of uniform leverage constraints can increase bank credit provision during economic downturns, in line with a precautionary cash holdings mechanism. Given the binding nature of the SLR, the relaxation of this constraint may be more effective than other countercyclical measures in allowing banks to extend credit.
Bank Investments in Venture Capital and Subsequent M&A Advisory Services
I examine the relationship between bank venture capital investments and subsequent M&A advisory services. The literature suggests that banks are strategic investors seeking complementarities between their different divisions. I find evidence that banks use venture capital investments as a way to build future M&A advisory relationships. I show that there is a 30% increase in the probability of being an M&A advisor conditional on investing in a company in the VC market. I find that banks make investments in sectors which have relatively high debt levels, possibly due to inter-temporal cross-selling opportunities. In line with prior literature, I show that banks benefit from relationships built at the VC stage through higher fees charged to the target companies in the subsequent M&A transaction. This is consistent with a certification role that the bank plays. This paper adds to the debate on the benefits and drawbacks of bank cross-selling activities and universal banking.
Banks' Next Top Model
I study the design of regulation using banks’ internal risk models. Specifically, I explore the optimal combination of capital requirements and penalties to ensure truthful reporting. I first characterize optimal regulatory capital and penalties when banks have private information about their risk. I find that heterogeneous penalties in practice could be rationalized provided sufficient variation in banks' preferences. I then use hand-collected data on risk model revisions to show that actual penalties provide only weak incentives for model improvements and fail to deter misreporting. In addition, my model suggests that recent changes in regulation may further impair truthful disclosure.
Belief Polarization, Information Bias, and Financial Markets
This paper studies how belief polarization affects financial markets. I develop an equilibrium model with two groups of investors whose polarized views are driven by biased private signals. Investors trade competitively in the market based on public information revealed by the equilibrium asset price and private information accumulated through word-of-mouth communication. Investors’ unconscious biases lead to belief divergence and generate excess volatility and trading volume. The information-sharing process further amplifies these effects. The public asset price does not fully eliminate investors’ unconscious biases.
How do Funds Deviate from Benchmarks? Evidence from MSCI's Inclusion of Chinese A-shares
An increasing amount of assets is managed by benchmark-tracking investment funds. This study investigates how benchmarking changes affect portfolio compositions in the cross-section of different investor types and stock characteristics. To that end, we exploit the phased introduction of Chinese A-shares to the MSCI Emerging Markets index, which was announced in June 2017 and implemented over the period from May 2018 to November 2019. This change presents a rare opportunity to estimate the impact of index changes and to shed light on cross-sectional implications. We document that particularly passive funds systematically deviate from the benchmark. Market capitalization, stock liquidity and stock volatility affect how benchmark changes translate to portfolio adjustments of mutual funds and ETFs. We then study how the changes in benchmark weights affect financial market outcomes, more specifically the comovement of returns. We find that these characteristics moderate the impact of benchmarking changes on financial market outcomes, suggesting that deviations from benchmarks matter.
Both Red and Green? Value Impact of Political Connections and CSR in China’s Cross-Border M&A
China’s communist history and special institutional setting constitute an ideal setting to study the role of political links on the relation between corporate social responsibility (CSR) and cross-border mergers and acquisitions (CBMAs) both encouraged by Chinese government. Using a sample of CBMAs attempted by Chinese listed firms between 2010 and 2018, we find that bidders with higher CSR performance achieve worse announcement stock returns, supporting the shareholder expense theory of CSR. The possible reason for CSR being at the shareholder costs is that such CSR is driven by incumbent CEO’s or Chair’s achieved political links, which means they are more likely to meet the Party and the government expectations instead of all stakeholders' benefits. In addition, we document that individual CEO’s ascribed political links could offset the negative effect of CSR on announcement stock returns.
Capital Regulation, Monetary Policy, and the Renegotiation of International Loans
We analyze which macroeconomic factors cause international lenders to drop out of syndicated loans. Increases in capital requirements in the lender country and decreases in borrower country policy rates imply a greater likelihood that foreign lenders will stop supplying capital in international syndicated loans. These results are robust to the inclusion of borrower country, lender country, and borrower round fixed effects. Using lender country capital regulations as instruments, we find evidence of significant economic spillover effects as international lender exits imply smaller loan amounts and shorter maturities.
Clear(ed) Decision: The Effect of Central Clearing on Firms' Financing Decision
Does credit derivative market regulation affect real economic outcomes? We investigate this question in the setting of the central counterparty (CCP) clearing reform on the corporate credit default swap (CDS) market. Exploiting the staggered introduction of CCP clearing to CDS contracts -- an insurance against firm default -- we uncover adverse real economic consequences. Firms whose CDS contracts are eligible for clearing with the monopolist CCP lose debt market funding, shrink their balance sheet, cut investment and become less profitable. As a response to the funding short-fall on debt markets, firms increase demand for bank loans. We theoretically motivate two channels through which the CCP environment can adversely affect firms’ debt funding situation: the hedging channel -- higher trading costs on the centrally cleared derivative market push hedged investors away from affected firms; and the arbitrage channel -- lower counterparty risk on the centrally cleared derivative market attracts investors from the bond market to the CDS market. Our results indicate that the arbitrage channel dominates the hedging channel.
Climate Change-Related Regulatory Risks and Bank Lending
We identify the effect of climate change-related regulatory risks on credit reallocation.
Our evidence suggests that effects depend borrower's region. Following an increase in salience of regulatory risks, banks reallocate credit to US firms that could be negatively impacted by regulatory interventions. Conversely, in Europe, banks lend more to firms that could benefit from environmental regulation. The effect is moderated by banks' own loan portfolio composition. Banks with a portfolio tilted towards firms that could be negatively affected by environmental policies increasingly support these firms. Overall, our results indicate that financial implications of regulation associated with climate change appear to be the main drivers of banks’ behavior.
Color, Loan Approval, and Crimes: The Dark Side of Mortgage Market Deregulation
This paper documents that racial differences in credit distribution during a general mortgage credit expansion can lead to unintended negative consequences on crime. Exploiting a federal mortgage market deregulation, we find a significant increase in mortgage approval to white borrowers, while the approval rate to black borrowers is unchanged. More importantly, the local housing boom induced by this credit expansion leads to an increase in money-related crime rates of black offenders. The results highlight an unintended adverse consequence of credit expansion on the welfare of the minorities.
Common Institutional Ownership and Corporate Carbon Emissions
In this paper, we examine whether common institutional ownership fosters collaboration among firms within the same industry in mitigating climate change, or reducing carbon emissions. We obtain strong and robust evidence that firms with more industry peers that are commonly owned by institutional investors have lower carbon emissions. In addition, we find a threshold exists that the impact on carbon emissions only holds when firms are commonly connected with a substantial number of peers. Overall, our results highlight the role played by institutional investors in tackling climate issues, with important implications for both climate- and antitrust-related regulations.
Strength in Differences: How Racial Integration Shapes Household Financial Decision-Making
Using proprietary geocoded data from the Panel Study of Income Dynamics, I examine whether local racial integration influences financial decision-making. I find that individuals residing in racially integrated communities are more likely to invest in public equity markets after accounting for individual and county-level differences. Exploiting within-individual variation from relocation as well as using Great Migration population shocks as an instrument, I demonstrate that integration has a causal effect on participation. Evidence suggests that racial integration improves local information quality, lowering informational barriers to participation. Moreover, this informational advantage enables integrated investors to achieve superior risk-adjusted performance on their local portfolios.
New Competition and Credit Rating Quality Inconsistency in the RMBS Market
We empirically investigate the impact of competition on rating quality in the credit rating market for residential mortgage-backed securities in the period 2017-2020. We find that small credit rating agencies (CRAs) − Dominion Bond Rating Service Morningstar (DBRS) and Kroll Bond Rating Agency (KBRA) − react differently to competitive pressures of its larger peers. DBRS assigns on average stricter ratings when facing higher competitive pressures of Moody’s, while KBRA has the tendency to provide more optimistic ratings. We also find that small CRAs tend to loosen its rating standards when competition of their larger peers increases, especially from Moody’s. Finally, we show that small CRAs are sensitive to issuers’ power as they provide more optimistic ratings on average when dealing with powerful issuers. Our findings suggest that a regulatory environment that stimulates new CRAs to enter the market does not necessarily increase the quality of credit ratings.
Creditor Control Rights and the Pricing of Private Loans
This paper investigates the influence of creditor control rights on the pricing of corporate loans. We construct a novel dataset, which combines individual borrower, lender, and loan characteristics with covenant violation data. Applying a standard quasi-regression discontinuity design on this data shows that creditors exploit their control rights to overprice new loans. Our dataset contains observations, in which borrowers are in violation only with some of their multiple creditors. This structure allows us to overcome the endogeneity concerns of our approach. In addition, we uncover novel cross-sectional and time-serial loan pricing patterns that can be explained by creditor control rights.
Creditors Control of Environmental Activity
I study the effect of creditor control on corporate environmental policy. Approximately 20% of loan agreements include clauses that grant lenders access to information about the borrowers' environmental profile. Such clauses are more prevalent among borrowers that violate a financial covenant, consistent with lenders heightening monitoring intensity on environmental behavior when their control rights increase. Following a covenant violation, firms reduce toxic chemical waste managed by 8-11%. The reduction stems from an increase in abatement initiatives rather than a reduction in production. Overall, my findings show that creditors use state-contingent contracts to shape corporate environmental policy.
Cross-Ownership and Corporate Debt Structure
This paper investigates the relationship between the borrowing firm’s cross-ownership and its choice between bank loan and public bond when raising new debt capital. We find that cross-ownership significantly reduces the firm’s usage of a bank loan when making debt issuance decision. The evidence from a quasi-natural experiment based on financial institution mergers mitigates concerns of reverse causality. Furthermore, the reduction in the likelihood to issue bank loan is more pronounced for firms with greater governance externality and information asymmetry. These findings highlight that the governance and informational roles of cross-ownership have real effects on corporate debt structure.
Cybersecurity and Financial Stability
We model how cyber attacks can impair banks’ operations and precipitate bank runs. Banks can defend themselves by investing in cybersecurity. But when banks share infrastructure (e.g. cloud-based platforms), cybersecurity is a weaker-link public good – the ability to thwart an attack is shaped by banks with the smallest benefit-to-cost ratios. In balancing private run risk and social cyber risk, banks underinvest in cybersecurity. This underinvestment is exacerbated by greater bank heterogeneity. Cyber audits and transfer payments between banks can facilitate socially optimal investment. We show how cybersecurity investments are shaped by bank characteristics and industry initiatives such as Sheltered Harbor.
Debt and Stock Market Participation
Inconsistent with economic theory that suggests almost all individuals should have exposure to the stock market, only about half of U.S. adults invest in equities through a brokerage and/or retirement account. Understanding the determinants stock market participation is important since equity ownership has implications for wealth inequality. A primary determinant of stock market participation is wealth, which proxies for participation costs. Most empirical tests of stock market participation employ either wealth (assets less debt) or assets as proxies for participation costs. The former implicitly assumes participation is equally impacted by a $1 increase in assets and a $1 decrease in debt (i.e., βWealth implies βAssets = -βDebt). The latter implicitly assumes debt has zero impact on participation (i.e., βDebt = 0). In this study, I empirically test these assumptions. I find that debt has a significantly larger impact on stock market participation than assets. Debt helps explain stock market participation even among the wealthiest households. My results are consistent with debt capturing behavioral factors (e.g., impulsivity and moral licensing) in addition to participation costs. My findings imply that future researchers should partition wealth into assets and debt as each capture unique aspects of the stock market participation puzzle. My findings also imply that policies which increase financial education may lead to more widespread stock market participation as financial education could decrease impulsive spending and increase equity investment. Such policies could help narrow the wealth inequality gap.
Debt Contract Enforcement and Product Innovation: Evidence from a Legal Reform in India
Due to a legal challenge, there was staggered introduction of fast track debt recovery tribunals (DRTs) across the states of India in the 1990s. Exploiting this plausibly exogenous variation in the efficiency of debt contract enforcement and using detailed information on product lines produced by the manufacturing firms, we study the causal effect of debt contract enforcement on product growth. We find that DRTs account for over 15% of the observed increase in firms' product scope during our sample period. Firms enter into new product lines in industries outside of their current scope of operation suggesting bolder innovation moves in response to DRTs. This increase in product scope is driven by firms in the top quartile of tangible asset distribution. These firms increase their borrowings and investments in R&D, plant and machinery, and selling & distribution expenses. There is also a significant improvement in their performance as measured by sales, profitability, and exports. In contrast, low tangible asset firms lose market share and experience a decline in their performance. DRTs also increase the aggregate state-industry level TFP by 6% driven by a significant increase in the TFP of the high tangible asset firms.
Directors' Personal Experience and Corporate Environmental Performance
We examine how directors' natural disaster-related experience affects a firm's environmental performance. We find that after a firm's directors experience environmental shocks at an interlocking firm, the focal firm improves its environmental performance in the following years. The impact is stronger when the disaster is more salient, then the focal firm has more interlocking directors with firms in disaster area, when these directors are more senior, and when the focal firm has a higher ratio of female directors. The effects are primarily concentrated in firms operating in environmentally sensitive industries, firms with an environmental committee, and firms with less financial constraint. We also find stronger evidence in firms located in areas with an eco-friendly inclination, i.e., counties with higher climate change beliefs or higher percentage of Democrat votes. Our results are not driven by peer effects and are robust to alternative environmental performance measures. Overall, we show that extreme weather events are likely to affect a firm's executives and directors' attitude towards climate change risk and that such exogenous shock can generate positive externality for interlocking firms' sustainability policy.
Discriminatory Versus Uniform Auctions : Evidence From JGB Market
In 2007, the Japanese government changed the format of auctions for 30-year Japanese government bonds (JGB) from uniform to discriminatory. We examine data before and after this change to assess whether this has lowered the borrowing costs of the Japanese government, in the largest government bond market in the world. As Ausubel et al. (2014) described, the general revenue ranking of uniform and discriminatory auctions is an empirical question. Our empirical result shows that this policy change lowered borrowing costs. We also show that a discriminatory
auction lowers the borrowing costs when the value of the bidders to JGB tends to be symmetric,
which is consistent with the prediction of Ausubel et al. (2014).
Do Firms Cater to Corporate QE? Evidence from the Bank of Japan’s Corporate Bond Purchases during the COVID-19 Pandemic
The Federal Reserve and Bank of Japan corporate bond purchase programs in response to the COVID-19 crisis primarily target bonds with five years or less remaining to maturity. This paper documents evidence suggesting that firms in Japan, but not in the U.S., have catered to the maturity-specific demand shock by shifting the maturity of new bond issues. Most strikingly, in Japan, there is a large and disproportionate reduction in issuance of bonds maturing in seven years, a previously popular maturity just above the maturity eligibility criterion. I argue that Japanese results are consistent with heterogeneous firms facing a trade-off between the gain from shortening maturities to match the positive demand shock and the cost of deviating from their intrinsically optimal maturities. An analysis of simultaneous issuances of multiple-maturity bonds further supports the catering explanation. Thus, this paper documents a novel unintended effect of corporate quantitative easing (QE) and has important policy implications.
Do Firms Have A Preference Order While Repaying Lenders? Relationship vs Transaction Banking
Do firms prefer repaying a relationship lender over a transaction lender or vice versa? It is unclear whether a shock to aggregate default in the economy would show a higher default rate towards the informed relationship lenders or the uninformed arm's length lenders? A difference in differences analysis shows that firms are more likely to default on relationship lenders compared to arm's length lenders. Firms default even more on relationship lenders that have helped the firm in the past, indicating that relationship banking may create a soft budget constraint. This effect is observed for both public and private lenders. The findings are robust to alternate definitions of relationship banking and controlling for the outstanding loan amount.
Do Information Acquisition Costs Matter? The Effect of SEC EDGAR on Stock Anomalies
I estimate the costs of information acquisition and the extent to which they explain stock anomaly portfolio returns. The SEC’s staggered implementation of EDGAR from 1993 to 1996 greatly lowered the costs of acquiring accounting information. I study how this quasi-exogenous and staggered shock affects the profitability of anomaly portfolios. The EDGAR introduction lowers the average alphas for the accounting anomalies by 4.0% per year, explaining over one-half of their pre-EDGAR alphas. The effect is stronger for the accounting anomaly portfolios that require more up-to-date accounting information, and for those that consist of EDGAR filer stocks with less information available in the pre-EDGAR period. By contrast, alphas for the non-accounting anomalies remain unaffected. These results imply that the costs of information acquisition, which are usually neglected, can be as important as the transaction or short sale costs.
Do Investors Overreact to Managerial Tones?
Behavior finance literature uses overconfidence to explain overreaction. This paper exams this theory with firms' annual report. Using a machine learning technique, I conduct a decomposition of managerial tones into different aspects of firms' fundamentals. Surprisingly, for SP500 constituents, I find the market reacts more to signals which are more informative about future fundamentals, indicating that investors extract value-related information from signals. A simple counterfactual shows that investors overconfidence can only explain an insignificant part of the filing period buy-and-hold excess return, only 0.03 basis points. Cross-sectional analysis shows that overreaction is most salient for high market valuation firms and medium-size ones. The evidence from SP500 does not confirm the existence of overconfidence.
Do Minority Banks Matter?
This paper estimates the elasticity of minority credit supply to deposit shares of minority-owned banks. To generate exogenous variation in minority bank shares, I use an instrument based on within-county tract-level variation in exposure to the Community Reinvestment Act. Minority credit supplied declines by 37% for up to six years if a census tract loses the presence of a minority-owned bank. 1% increase in county deposit share of such census tracts leads to a 3% decrease in aggregate minority homeownership. I consider competing explanations and suggest that the findings are best explained due to the severing of minority banking relationships.
Do Offshore Activities Teleport Information: Evidence from Foreign Analysts’ Coverage of U.S. firms
We find that non-U.S. analysts are more likely to cover U.S. firms that have offshore activities in their domiciled countries than those without any offshore activities. Forecast accuracy is also improved as U.S. firms sell more goods and products in the analysts’ domiciled country. The effect of offshore activities is more pronounced when the institutional infrastructure of the target country is weaker and the information transparency of the underlying firm is poorer, supporting the conjecture of an information advantage arising from the offshore activities among the domiciled analysts. Consistent with the investor demand hypothesis, U.S. firm’s offshore activities are positively related to the search volume of the firm’s EDGAR filings and institutional equity holdings from the target countries. The study highlights that offshore activities can teleport an information advantage to a group of foreign investors and analysts due to the domestic presence of the U.S. firms through their offshore network.
Does Beauty Matter in Mutual Fund Managers’ Performance?
This paper examines the effect of mutual fund managers’ beauty on their performance. Using the data from China, we document that in both portfolio analysis and regression analysis, team-managed funds managed by attractive managers outperform other team-managed funds, while this phenomenon is absent among solo-managed funds due to the counteraction of better stock-picking skills and worse market timing skills for attractive solo managers. For channels of beauty premium, we find attractive fund managers have more opportunities to visit listed firms to get first-hand information and are more centralized in their social network, which enhances their information advantages. Additional evidence shows that while fund managers’ beauty does not attract more fund inflows, attractive managers are more likely to gain promotion even if they are not performing well. Overall, our study shows that physical attractiveness is a significant factor that determines mutual fund managers’ information acquisition and performance.
Does Competition Improve Information Quality: Evidence From the Security Analyst Market
This paper studies the effect of strategic interaction on the quality of information provided by experts. I estimate the incentives and the information structure of security analysts who compete to make earnings forecasts. Security analysts are rewarded for being more accurate than their peers, which creates competition. This reward for relative accuracy leads analysts to distort their forecasts to differentiate themselves, but also disciplines them to be less influenced by the prevailing optimism incentive. I structurally estimate a contest model with incomplete information that captures both effects, adapting the estimation of common value auctions using indirect inference. My model disentangles the payoff for relative accuracy from the payoffs for optimism and absolute accuracy.
Using the model, I conduct counterfactuals to evaluate policies that reduce the importance of relative accuracy in analysts‘ payoff. I simulate the effect of these policies on the quality of information. I find that the disciplinary effect dominates in the current market: the reward for relative accuracy reduces individual and consensus forecast errors by 33.29% and 58.45% respectively. Meanwhile, this improvement is at a cost of increasing individual and consensus forecast variances by 3.77% and 4.37% due to the distortionary effect. For each security, it is optimal to have moderate competition between the covering analysts, because more competition generates more aggregate information but also intensifies the distortionary effect.
Does Fintech Credit Reduce Income Inequality? Evidence from Migrant versus Native Business Owners
We examine whether Fintech-based microfinance reduces the income gap between the migrant and the native small business owners. Using the data on business owners registered with the largest Fintech firm in China, we find that with the access to microloans, migrants achieve greater business revenues compared to their native counterparts. We further find that the effect is more pronounced in the cities with higher concentration of migrants, and among business owners who operate their businesses either online or offline via QR code (but not both). The channel analysis supports the financial constraint hypothesis that the effect is more prominent in the businesses with less real estate collateral for financing. Finally, we document an incomeconverging effect between the migrant and the native business owners in cities with more developed mobile payment system. Overall, our findings support that Fintech playys an important role in reducing income gap between native and migrant small business owners.
Does Inflation Heterogeneity Matter in Household Financial Decisions? Evidence from the Mortgage Market
This paper investigates the impact of inflation heterogeneity on household financial decisions. A parsimony model predicts low income households, who experience relatively higher inflation, have the incentives to reduce their holdings of nominal assets, move their portfolios towards real assets, and reduce their saving rates. The actions can be mapped into household mortgage takings. Consistent with the model, I find the bottom (top) income households increase (decrease) mortgage takings when the inflation gap is large. To establish identification, I use the Chinese Yuan appreciation relative to the US Dollar as an instrumental variable of US inflation heterogeneity, because low income households have higher expenditure exposures to the price of tradable goods. I also use the July 21 2005 Chinese Yuan reform as an event study and the same pattern holds. The results are neither driven by an income effect through local China trade exposure nor by a credit supply effect.
Does Social Interaction Spread Fear among Institutional Investors? Evidence from COVID-19
We study how social connectedness affected mutual fund manager trading behavior in the first half of 2020. In the first quarter during which the COVID outbreak occurred, fund managers located in or socially connected to COVID hotspots sold more stock holdings compared to a control group of unconnected managers. The economic impact of social connectedness on stock holdings was comparable to that of COVID hotspots and was elevated among “epicenter” stocks most susceptible to the pandemic shock. In the second quarter, social interaction had an overall negative effect on fund performance, but this effect depended on manager skill; unskilled managers who were connected to the hotspots underperformed, while skillful managers suffered no deleterious effect. Our evidence suggests that social connections can intensify salience bias for all but the most skilled institutional investors, and policy makers should be wary of the destabilizing role of social networks during market downturns.
Doing Good and Doing It With (Investment) Style
We study the asset allocation, spending behavior, fees, and risk-adjusted performance of U.S. private foundations from 1991 to 2016. We find that large foundations outperform and generate risk-adjusted returns of about one percent per year. We document considerable time series variation in alphas and weakening performance persistence. Because of the constraints imposed by the five percent spending rule and accommodating monetary policy, private foundations also increase risk-taking and reach for yield. Due to these constraints, a conservative asset allocation will decrease real principal balances over time resulting in less charitable giving. In an infinite horizon setting, we show that foundations can maximize the present value of future distributions under alternative spending rules.
Economic Narratives and Market Outcomes: A Semi-supervised Topic Modeling Approach
We employ sLDA to extract the narratives discussed by Shiller (2019) from 7 million NYT articles over 150 years. The estimation addresses look-ahead bias and changes in semantics. Panic and the narrative index positively predict market return and negatively predict volatility. Panic presents time-varying risk aversion. The narrative predictability increases recently at both market and portfolio and monthly and daily intervals. The narrative index constructed from 2 million WSJ articles over 130 years retains its predictive power, but Stock Bubble emerges as a negative market predictor. Media customizes their narratives to their readers, having a diverse effect on the market.
Effects of Bank Capital Requirements on Lending by Banks and Non-Bank Financial Institutions
In this paper, we empirically investigate effects of bank capital requirements on the lending activity of banks and non-bank financial institutions (NBFIs). As a quasi-natural experiment, we exploit a sudden and sizeable increase in bank capital requirements imposed by the European Banking Authority (EBA) within the framework of its capital exercise in 2011. The exercise affected some but not all of the German banks and did not have a direct impact on NBFIs. Such implementation creates an ideal setting to apply a difference-in-differences methodology to the data from the German credit register. We find that, following the capital exercise, NBFIs and non-EBA banks slowdown their real sector lending less, compared to the EBA banks, by extra 2.2% and 1.6% per quarter, respectively. This effect is observed for several categories of NBFIs: insurance companies, financial enterprises, and financial services institutions excluding leasing companies. However, the heterogeneity of the effect across different NBFI categories requires further investigation. In order to identify the underlying mechanism of the credit reallocation and to assess possibilities of regulatory arbitrage, we closely examine a link between banks and NBFIs. This aspect has a very limited coverage in the previous literature and is particularly important in the environment of low interest rates, which could intensify the competition between commercial banks and NBFIs seeking positive yields. Our work examines the spillover effects of the banking regulation and the growing importance of less regulated NBFIs. Our results contribute to the assessment of the impact of bank capital requirements on the distribution of risks in the system and on the overall financial stability.
Employee Discrimination and Corporate Morale: Evidence from the Equal Employment Opportunity Commission
Using a difference-in-differences model around Equal Employment Opportunity Commission (EEOC) discrimination announcements and Glassdoor.com employee reviews suggest a negative causal link between discrimination publicity and employee morale. An EEOC discrimination announcement results in a 9.08% and 3.90% decrease for employee approval of a CEO and firm, respectively. The effect clusters in firms headquartered in the southern U.S. and that are in service orientated industries suggesting taste-based discrimination. These announcements do not impact stock returns, sales, or CEO turnover but lead to human capital risk, workforce reductions, and decreases in Tobin's Q implying a discrimination, instead of litigation, effect.
Estimating and Forecasting Long-Horizon Dollar Return Skewness
We develop a parametric estimator of the physical skewness of an asset's discrete ("dollar") return over long horizons from the assumption that the asset's value can be modelled using a stochastic process from the affine stochastic volatility (ASV) model class. Taking compounding and leverage effects into account, we demonstrate that our estimator is close to unbiased and efficient, setting it apart from other recent estimators. In a further contrast to those other estimators, it also lends itself naturally to forecasting skewness. Applying our estimator to representative stock indexes, we show that the skewness of long-horizon dollar returns is far less extreme than suggested in the current literature.
Explaining Greenium in a Macro-Finance Integrated Assessment Model
How do firms' environmental performances affect cross-sectional expected stock returns? Using a third-party ESG score, I find that greener stocks have lower expected returns. This greenium remains significant after controlling for systematic and idiosyncratic risks. Green stocks hedge climate-related disasters, contributing to the greenium. A macro-finance integrated assessment model featuring time-varying climate damage intensity, recursive preferences, and investment frictions quantitatively explains the empirical findings. The model implies a positive covariance between climate damages and consumption, which justifies a high discount rate and a low present value of carbon emission.
Outsourcing Workplace Safety
I study if firms deliberately sacrifice workplace safety for profits by using contract workers, for whom they are not legally liable. I exploit a regression discontinuity design around the amendment to the Occupational Health and Safety Act in Korea, 2017, which expanded the legal accountability of firms to cover contract workers. The number of contract workers decreased by 18.1% in affected establishments compared to unaffected establishments. This change was not compensated by direct hiring, causing overall employment to fall by 1.3%. Working hours and wage costs paid to directly hired employees increased to make up for the resulting losses in work hours from the contract workers. Workplace safety improved at affected establishments at the cost of higher safety investment. Profitability dropped in affected firms, and those firms reacted by shrinking investments. The results are consistent with firms strategically outsourcing risky jobs to contract workers to offload their duties on workplace safety.
Bank Presence and Health
What role does bank presence play in improving health? To explore this question, I use a policy of the Reserve Bank of India from 2005 that incentivizes banks to set up new branches in underbanked districts, defined as having a population-to-branch ratio larger than the national average. In a regression discontinuity design, I compare households in districts just above and just below the national average. Six years after the policy introduction, households in treatment districts are a third less likely to be affected by an illness in a month. They miss fewer days of work or school due to an illness and have lower medical expenses. Ten years after the policy was introduced, I observe persistently lower morbidity rates, higher vaccination rates, and lower risks associated with pregnancies. I provide evidence that two previously understudied aspects of banking contribute to the effect: households gain access to health insurance and health care providers gain access to credit. In equilibrium, I observe an increase in health care demand and supply.
Flag Tag: Credit File Disaster Flags As Social Insurance Tags
This paper finds 59.2 million people had a ‘disaster flag’ on their US credit file (2010 - 2020) with broad geographical use during the COVID-19 pandemic. Disaster flags mask adverse credit file data with the aim of protecting credit access following disasters such as hurricanes and wildfires. Flags are voluntarily applied by lenders to borrowers’ credit files. I describe the selection of lenders and borrowers into applying these flags over twenty years and estimate the effects of flags on credit access using a difference-in-difference design. There is adverse selection into flag use: people using flags are ex-ante riskier and defaults masked by flags are riskier than non-flagged defaults. I find small average effects of flags on credit scores (1.5 - 2 pp) driven by larger (10 - 15 pp), temporary effects for those with pre-disaster defaults or subprime credit scores. Finally, the paper considers a counterfactual social insurance regime automatically masking all new defaults during natural disasters and finds doing so would have limited predictive loss.
Following the Crowd: Anomalies and Crowding by Institutional Investors
This paper investigates the relation between crowded trades, those in which many investors hold the same stocks possibly exhausting their liquidity provision, and institutional investors' trading activity on a set of twelve well-known stock market anomalies. Consistent with previous studies, we find little evidence of overlap among institutional investors’ portfolios, however, we observe significant crowdedness at the security level, especially among larger stocks. We select days-ADV as our preferred crowding measure. Days-ADV can be interpreted as how many days, based on the daily average turnover over the previous quarter, would it take for institutional investors to exit their collective position in a given security. We show that a high-minus-low crowding portfolio delivers economical and statistically significant excess returns, and its variation is distinct from other traditional risk factors. We extend our analysis to our set of stock market anomalies and find that anomaly risk-adjusted returns appear to be concentrated among the most (least) crowded stocks for the long-leg (short-leg) portfolio. This finding is consistent across all the anomalies in our sample and remains significant after publication dates. Additionally, we find that our results are stronger among holdings of transient institutions. We hypothesize that crowded equity positions in anomaly stocks increase institutional investor’s exposure to crash risk and fire sales, which adds a new consideration to the limits of arbitrage.
Fund Flows in the Shadow of Stock Trading Regulation
Trading suspension, a widely adopted regulatory rule, prevents information from being incorporated into stock prices. Using a sample of 3,205 long-lasting suspension events between 2004–2018, we show that mutual funds holding suspended stocks generally fail to adjust for stale prices, generating stale net asset values (NAVs). We find that investors exploit predictable fund performance that realizes quickly after trading resumes: flows positively respond to firm-specific news about suspended stocks in fund portfolios. Portfolio disclosure plays a key informational role in distorting flows. Our findings suggest that regulatory interventions on trading activities can create negative externalities among mutual fund investors.
New technologies are a main driver of economic growth, necessary for the economy to transition to clean energy solutions. Startups can be key developers of cleantech innovations that will be crucial for the pursuit of both economic growth and climate standards. Successful startups rely mostly on Venture Capital (VC) financing, yet it is unclear whether typical VCs' characteristic, such as targeted returns and exit horizons, are in line with this type of innovation. This paper investigates this hypothesis and finds that: 1) on average green startups receive their rst round 6 months later than other startups, and 2) green startups are 6.2% less likely to be acquired. The results suggest that VCs might not be the best financiers of cleantech startups and this might slow down the production of clean energy solutions.
Help Your Employees, Help Your Firm:Evidence from U.S. Paid Sick Leave Mandates
This paper exploits the staggered implementation of state-level paid sick leave (PSL) mandates to assess their real effects on U.S. corporations. We find that employees’ better access to sick pay leads to higher firm productivity and profitability. First, we show that the positive effects on performance are more pronounced for firms with more expensive labor. The results suggest that employees who prefer sick pay to incremental pecuniary compensation will have an incentive to exert more efforts, resulting in better firm performance (incentive channel). Second, we show that the generosity of PSL boosts firm performance by improving employee health. To this end, we find that the performance improvement is stronger for firms operating in industries which tend to require physical presence in the workplace (health channel). Moreover, our main results are largely driven by firms headquartered in counties with higher social capital, which are less prone to moral hazard stemming from PSL. Finally, additional tests reveal that the increased PSL coverage is associated with higher firm value and greater use of leverage. Overall, our paper demonstrates a Pareto improvement resulting from the provision of fringe benefits, and it contributes to the recent debate on the effectiveness and efficacy of PSL during the COVID-19 crisis.
Heterogeneous CSR approaches
Theoretical CSR literature argues that firms approach CSR via either strategic CSR, CSR-as-insurance or corporate greenwashing. Where empirical CSR literature primarily analyses CSR in general, we show that it is precisely the heterogeneity in CSR approaches that shapes the societal contribution and financial performance of firms. Using a novel method which segregates the promised to realised CSR performance of firms, we find that firms approach strategic CSR, CSR-as-insurance and corporate greenwashing respectively 50%, 24% and 26% of the time for a global sample. By comparing the societal contribution and financial performance of firms, we show that contributing to societal welfare through strategic CSR enhances profitability, whereas corporate greenwashing simultaneously deteriorates societal welfare and firm value.
How do Investors Learn as Data Becomes Bigger? Evidence from a FinTech Platform
We study how investors learn from data, including their response to the availability of new predictive signals. We identify learning effects thanks to a panel of systematic trader performance outcomes from a FinTech platform that organizes trading contests under tightly controlled conditions. Investor outcomes improve with experience during both backtest and out-of-sample periods. The outcomes of experienced investors improve when additional predictive variables are introduced to the common part of their information sets. To explain these results, we model an investor as choosing a portfolio by learning from historical data while also taking model uncertainty into account. Estimating our model reveals that inexperienced investors tend to use fewer predictive signals, in keeping with the proposed mechanism.
How Institutional Dual-Holders Affect Companies: Evidence from U.S. Mutual Funds?
The divergence of objectives between shareholders and creditors can result in a conflict of interest and it is important to empirically understand how this conflict of interest affects company value and performance. In this research, I am going to provide evidence regarding shareholder-creditor conflict measured by dual-holding of U.S. mutual funds.
The estimations find the positive correlation between company value and the existence of dual-holders for distressed companies. The presence of dual-holders with block equity stakes (block-dual-holders) has a negative correlation with a company value showing higher magnitude for companies far from distress. However, the level of equity and bond stakes of block-dual-holders is associated with a positive impact on the value of distressed companies. Also, the findings show that the presence of dual-holders in distressed companies positively correlates with company value for financially constrained entities, showing that dual-holders mitigate shareholder-creditor conflict around distress alleviating concerns related to access to external financing. The presence of block-dual-holders across financially constrained companies is associated with a positive influence on company value. This impact decreases when I control for the presence of other dual-holders without block equity stakes.
In addition, I propose several alternative approaches to address the endogeneity concerns and bring additional credibility to my findings. However, at this stage, the working paper includes explanations of potential algorithms and equations only.
This research will help to understand whether dual-holders can alleviate shareholder-creditor conflict of interest from many perspectives, e.g. size of investor stakes, their monitoring activity, trading strategy and connectedness. It will bring several useful insights for researchers, industry professionals, institutions and policymakers.
How Many Female Seats on a Board? Board Gender-Diversification, Power, Risk-Taking, and Financial Performance
Using a novel combination of empirical tools and analyses, we demonstrate that if a female director is unlikely to have any personal power or influence on the board, her addition to the board will have no significant impact on firm risk-taking and performance. However, with increasing power/influence on the board (via greater numerical strength or non-token aggregate position), female directors will tend to reduce the excessive risk-taking behavior of the firm and, to the extent that the gender-diversification process is non-disruptive, the expected risk-reduction effect can feature significant increases in profitability and firm value. We also show that the increase in profitability is driven not by market timing of equity issues but by the sale of less productive physical assets, more retained earnings, paid-down debt, and less cash flow volatility. Overall, our results show that board gender diversity affects corporate risk-taking culture and firm performance in a value-maximizing manner, particularly when gender diversification is both non-tokenistic and non-disruptive.
Impact of Economic Shocks on Financial Access: Evidence from COVID-19 Pandemic
This paper examines the impact of an economic shock and the government response on financial access for underserved consumers. Using foot traffic to consumer lenders as a proxy for loan demand, we find that the shelter-in-place order, new COVID-19 cases, and the government relief program (PEUC) are associated with a drop in visits to consumer lenders after controlled for the online borrowing and the supply of credit. Using natural experiments of the statewide shelter-in-place order and FPUC program, we find that the lockdown suppresses financially underserved consumers' access to credit, while the supplemental paychecks (FPUC) cushion their economic blow by further reducing visits to consumer lenders. We also find that regular unemployment insurance is less effective in reducing demand for consumer credit in financially underserved areas than in metropolitan areas. The demand for consumer credit is positively correlated with the average consumption level in an area. Lastly, we find differences in the impact of the government relief programs on visits to banks and visits to consumer lenders.
Informative Covariates, False Discoveries and Mutual Fund Performance
We present a novel multiple hypothesis testing framework for selecting outperforming mutual funds, named the functional False Discovery Rate “plus”. Our method incorporates informative covariates in estimating the False Discovery Rate. It gains considerable power (up to 30%) in simulations over the Barras–Scaillet–Wermers (BSW) approach. We show that portfolios based on four new informative covariates and five well-known ones demonstrate truly positive performance and surpass the BSW portfolios and those based on sorted covariates. We note that all covariates carry valuable information in mutual fund selection that is persistent, even over the recent period.
Institutional Trading around FOMC Meetings: Evidence of Fed Leaks
Fed leaks to the financial sector are actively exploited by institutional investors to trade ahead of the Federal Open Market Committee (FOMC) meetings. Using detailed transaction records from Ancerno, I find evidence consistent with informed institutional trading on the stock market on the days before FOMC scheduled announcements. The institutional trading imbalance on highly exposed stocks is in the same direction of the subsequent monetary policy surprise. The magnitude of this result is economically significant. I find that trades in anticipation of FOMC meetings are particularly strong before easing monetary policy shocks - when the aggregate market reaction is positive -, for the most-active traders, and for the hedge funds that are headquartered close to one of the regional reserve banks. Fed informal communication with the financial sector seems to be driven by the non-voting members of the Federal Open Market Committee. These findings contribute to an information-based explanation of the pre-FOMC drift and, from a policy perspective, suggest that any benefits of Fed unofficial communication must be balanced against the risk of giving some investors an unfair advantage.
Intellectual Property Rights and Employee Stock Option Compensation: Evidence from Court of Appeals Federal Circuit Ruling in 2008
This study uses the Court of Appeals Federal Circuit (CAFC) ruling in 2008 as a quasi–natural
experiment to examine the effects of the patent ownership shift from inventor employee to an
employer on employee stock option compensation and its consequences on a firm’s innovation
activities. I find that treated firms, which are located in formerly pro-employee invention
assignment states, increase employee stock option compensation and innovation activities
following the CAFC ruling in 2008. Main results are not driven by global financial crisis or
firm’s financial constraint. My evidence highlights the role of employee option compensation
in motivating employees’ innovation activities effectively.
Interest Rate Risk, Prepayment Risk and Banks’ Securitization of Mortgages
This paper shows the importance of interest rate risk and prepayment risk in fixed-rate mortgages in influencing banks’ securitization of mortgages. Banks with longer-maturity liabilities are more capable of taking the interest rate risk and therefore securitize fewer mortgages. In contrast, banks with shorter-maturity liabilities securitize more mortgages and originate fewer jumbo mortgages, which can not be securitized through Fannie Mae and Freddie Mac. Moreover, household mortgage refinancing induces prepayment risk. The prepayment risk matters more for banks with longer maturity liabilities, due to their high retention of mortgages on balance sheets. Ex ante, anticipating the prepayment risk, banks with longermaturity liabilities securitize more mortgages. Ex post, banks with longer-maturity liabilities are less likely to help households refinance their existing mortgages.
Investor (Mis)reaction, Biased Beliefs, and the Mispricing Cycle
We construct a new measure that captures the disparity between the market reaction to earnings information and the earnings surprise ("Return-Earnings Gap", "REG"). High REG scores positively predict analyst forecast errors and firm mispricing (overvaluation) scores, especially for build-up anomalies. Analyst forecast errors are slower to converge when REG provides confirming information. In turn, REG is positively predicted by analyst forecasts errors and higher mispricing, leading to a continuation of firm overvaluation over a few quarters. Overall, our results reveal how the market’s (mis)reaction feeds back into the belief formation of analysts, which partially explains the slow correction of firm mispricing.
Is Capital Reallocation Really Procyclical?
Aggregate reallocation is procyclical. This empirical observation is puzzling given the documented fact that the benefits to reallocation are countercyclical. I show that this procyclicality is entirely driven by reallocation of bundled capital, which is highly correlated with market valuation and bears no consistent relation to measures of productivity dispersion. Reallocation of unbundled capital, on the contrary, is countercyclical and highly correlated with dispersion in productivity growth, both within industry and across industries. To rationalize these facts, I propose a heterogeneous agent model of investment featuring two distinct used-capital markets and a sentiment component. In equilibrium, unbundled capital is reallocated for productivity gains only, whereas bundled capital is also reallocated for real, or perceived synergies in the equity market. While equity overvaluation negatively affect total factor productivity (TFP) by encouraging excessive trading of capital, such an adverse impact is largely offset by eased frictions to reallocation in the unbundled capital market.
Is China's Belt and Road Initiative a Zero-Sum Game?
Extant literature finds that foreign infrastructure investments tend to increase cross-border economic activity between investor and recipient countries. We question whether such an increase comes at the expense of trade with third-party countries (a “zero-sum hypothesis”), or whether the infrastructure investment leads to an increase in overall trade (a “lifting all boats hypothesis”). Our investigation is within the context of the Chinese Belt and Road Initiative (BRI). In a sample spanning 2013 to 2018 and covering 1,135 BRI projects in 110 countries, we find strong evidence in support of the zero-sum hypothesis. The increase in cross-border economic activity (imports, exports, and M&A flows) with China is accompanied by a decrease in activity with third party countries. Further, we show that, following BRI investments, BRI countries trade more with other countries that are politically aligned with China, but less with countries that have recently been visited by the Dalai Lama. Overall, our evidence points to both a “zero-sum” nature of the impact of infrastructure on cross-border trade, and to the existence of a BRI “network” that favors countries that are politically aligned with China.
"It's Not You, It's Them": Industry Spillovers and Loan Portfolio Optimization
Consistent with banks internalizing industry spillovers arising from product market competition, I show that lenders with an industry-wide exposure extend loans with stricter covenants to these firms. This allows these lenders to curb the growth appetite of borrowers competing in the same market, thereby maximizing the lenders’ loan portfolio value at the industry level. Specifically, I find that these lenders impose stricter capital-based covenants, which deters debt-funded growth and excessive risk-taking “ex-ante” by requiring more “skin-in-the-game”. They are also more likely to impose capex restrictions and less likely to include payout restrictions. This behavior is more pronounced for borrowers in more mature industries with less market growth potential. Exploiting bank mergers, I verify that these findings are robust to endogeneity concerns or alternative explanations.
Leasing as a Mitigation of Financial Accelerator Effects
We document that leased capital accounts for about 20% of the total physical productive assets used by U.S. public listed firms, and its proportion is more than 40% among small and financially constrained firms. Leased capital ratio exhibits strong counter-cyclical pattern over business cycles and positive correlation with aggregate uncertainty. In this paper, we argue that leasing has important mitigation effects for the financial accelerator mechanism. We explicitly introduce buy versus lease decision into the Bernanke-Gertler-Gilchrist financial accelerator model setting to demonstrate a novel economic mechanism: the increased usage of leased capital when financial constraints become tighter in bad states mitigates the financial accelerator mechanism and thus the response of macroeconomic variables to negative TFP shocks and positive uncertainty shocks. We provide strong empirical evidence to support our mechanism.
Manager Uncertainty and the Cross-Section of Stock Returns
This paper evidences the explanatory power of managers’ uncertainty for cross-sectional stock returns. I introduce a novel measure of the degree of managers’ uncertain beliefs about future states: manager uncertainty (MU), defined as the count of the word “uncertainty” over the sum of the count of the word “uncertainty” and the count of the word “risk” in filings and conference calls. I find that managers’ level of uncertainty reveals valuation information about real options and thereby has significantly negative explanatory power for cross-sectional stock returns. Beyond existing market-based uncertainty measures, the manager uncertainty measure has incremental pricing power by capturing information frictions between managers’ reported uncertainty and investors’ perception of uncertainty. Moreover, a short-long portfolio sorted by manager uncertainty has a significantly positive premium and cannot be spanned by existing factor models. An application on COVID-19 uncertainty shows consistent results.
Managing Climate Change Risks: Sea Level Rise and Mergers and Acquisitions
Using a large sample over the period 1986 to 2017, we show that companies with higher exposure to climate change risk induced by sea-level rise (SLR) tend to acquire firms that are unlikely to be directly affected by SLR. We find that acquirers with higher SLR exposure experience significantly higher announcement-period abnormal stock returns. Post-merger, analyst forecasts become more accurate and environmental-related as well as overall ESG scores improve.
Mandatory Central Clearing and Financial Risk Exposure
I analyze the effect of mandatory counterparty default insurance (central clearing) of over-the-counter (OTC) derivatives on aggregate financial risk exposure. I carefully model the competitive mechanisms in both the OTC derivatives and their insurance market. I show that the introduction of mandatory insurance empowers the for-profit central counterparty (CCP) to raise prices, wherefore only larger clients opt to additionally insure their derivatives (lower credit risk). Smaller clients instead exit the market and remain unhedged (higher market risk). I conclude with a model calibration and counterfactual policy evaluation for the EuroDollar FX derivatives market, showing that mandatory insurance increases aggregate financial risk.
Mandatory Pension Saving and Homeownership
We explore the implications of mandatory minimum contributions to retirement saving accounts over the life cycle. Mandatory minimum contributions alter housing market entries and have substantial welfare effects. We propose a flexible retirement saving scheme that only requires individuals to contribute to tax-deferred accounts if they have not built up sufficient savings. This flexible retirement saving scheme partly alleviates the unintended side-effects of mandatory minimum contributions and simultaneously ensures that individuals build up sufficient retirement savings.
Market Liquidity after Banning Aggressive Proprietary Trading
There is an increasing concern that fast trading firms magnify adverse selection costs and illiquidity by picking off stale quotes. Will restrictions on aggressive fast trading improve market liquidity? This article investigates the liquidity effects of banning proprietary traders from liquidity taking on the Aquis Exchange from February 8, 2016. I find that while realized spreads increased on Aquis, effective spreads and price impact declined substantially in the first four weeks after the change. The results are consistent with the theoretical prediction that limiting fast traders from liquidity taking improves local liquidity.
Market Power, Innovation Flow and Macroeconomic Dynamics
The technology (patent) market is designed for firms to efficiently allocate innovations. This
paper argues that many patent buyers are large incumbents and their market power may generate misallocation in the technology market and thus affect aggregate productivity. Using data on patent assignment and citation from USPTO, I show that although patents have better average quality after transaction, those purchased by large firms with similar patents receive lower citations than their non-traded counterparts. Then I develop a general equilibrium growth model to study the trade-off between market power and productivity in technology market. Incumbents make two types of technology acquisitions defensive ones in which they buy incremental innovations to maintain market leadership and achieve minor quality improvement, and productive ones where the acquirer buys radical innovations for business expansion. The relative gains in these two types of tech transactions influence the inventors' choice over incremental and radical R&D, and thus affect aggregate productivity and social welfare. I use the calibrated model to show that as some productive firms accumulate market power, they pay generously in internal acquisitions, attracting inventors to do non-radical innovation. The misallocation in technology market slows down productivity and harms social welfare.
Media Coverage and The Cross-Section of Mutual Fund Herding
This paper shows that media coverage of fund holdings positively affects an average fund manager's herding behavior through information creation and dissemination roles. Our simple measure, called Media-Driven-Herd (MDH), captures a fund manager's tendency to herd due to media coverage. We find that low media-induced-herding funds outperform their peers by about 2.5% per year. Media is positively related to buy-herds, whereas negatively related to sell-herds. Managerial experience incentives that attenuate herding behavior are eroded by the media and managers herd in the direction of the news's informational content. Our evidence suggests that media coverage can exacerbate managerial herding behavior due to limited attention and flow catering and serves as channels that make herding effective.
Mind the Income Gap - Partial Hedging of Interest Rate Risk within Banks' Business Model
We implement a recently established approach to investigate the interest rate risk of banks with extensive engagement in maturity transformation. Therefore, we contribute to the emerging literature contradicting modern banking theory's view on interest rate risk as an inevitable consequence of banks' maturity mismatch. For our sample, we confirm an exposure of banks' net interest income to changing market rates.
We also find evidence for an alignment of banks' interest income and expense sensitivities which might indicate an implied interest rate risk hedge by their business model. Banks with lower expense sensitivities show significantly higher loan maturities in their balance sheets, especially if their difference between interest expense and income sensitivity is small. Our results shed light on an implicit hedging mechanism within banks' business models, its (in)completeness, the use of interest rate derivatives, and consequences for adequate regulation.
Municipal Bankruptcy and the Economic Costs of Financial Contagion
This paper examines whether one municipality’s bankruptcy exposes other local governments to economic costs of financial contagion. For identification, we exploit idiosyncratic bankruptcies occurring due to legal judgments, financial speculation, other financial mismanagement, and failed public projects.
Using a cross-border setting, we show that other local governments located in the state of the bankrupt municipality are less likely to issue debt over the following year. The negative effect on credit market access is transitory as it disappears in the subsequent years. To identify the economic consequences of the limited credit market access, we exploit ex-ante heterogeneity in local governments’ maturity of long-term debt within the state of the bankrupt municipality. We find that local governments with high fractions of maturing debt—therefore with a high immediate demand for credit—persistently cut their public expenditures in the three years following the bankruptcy. This effect is mainly driven by a decrease in capital outlays. The lower investments by local governments also transmit to the private sector. We find that tradable employment and establishments decrease in counties with high fractions of maturing debt.
Overall, our results highlight the importance of functioning municipal credit markets since even temporary credit market disruptions have a permanent adverse effect on the development of other local governments that rely on debt financing.
Municipal Finance During the COVID-19 Pandemic: Evidence from Government and Federal Reserve Interventions
We study the functioning of the municipal bond market during the COVID-19 pandemic. The average offering yield increases while the number of new issues drops when county-level COVID-19 case and death counts rise. Exploiting the differential timing of local policy actions, we find that emergency declarations lead to a 69 basis-point increase in offering yields and a significant drop in new issuance. Investors shun transportation and dedicated tax bonds or bonds issued in fiscally unhealthy states. The Federal Reserve's unprecedented interventions through two municipal liquidity facilities have calmed the market. The reopening of local economies has led to a significant drop in offering yields.
Does Culture Matter in Corporate Cash Holdings?
This article identifies national culture as an important factor affecting corporate cash holdings by using China and its national culture, Confucianism, as the setting. We find that firms located in regions with higher levels of Confucian culture hold persistently higher levels of cash. Next, we employ an instrumental variable to establish causality of the culture effect and the IV estimates show a compelling economic magnitude. The culture effect is stronger for more financially constrained and riskier firms, suggesting precautionary motives as the underlying mechanism. Besides, we find that the culture effect remains intact after controlling for corporate governance heterogeneity, which rules out the agency motives. Further, we provide additional evidence suggesting higher cash holdings is an efficient outcome. Finally, we also show that both the CEO/board chairman’s Confucian background and the firm’s headquarter cultural environment matter.
Natural Disasters and Bitcoin
The impact of cryptocurrency mining (specially Bitcoin) on climate change has been widely discussed, however, the reverse direction is quite neglected. The occurrence of natural disasters as repercussions of climate change halts the operations of cryptocurrency mining that further reduces or pauses the supply of a currency in the market on one side and affects the rewards earned by miners on the other side. Hence, the working conditions and efficiency of mining devices are important determinants of performance of the cryptocurrency mining industry. This is quite evident from the recent (April 2021) power cut due to flooding in one of the cryptocurrency mining hubs in China which then results in a decrease in the hash rate and price of Bitcoin (BTC) by approximately 20%. To the best of my knowledge, there is no study till date about the effect of calamities on cryptocurrencies. Therefore, my study fills this research gap and analyzes the impact of natural disaster on cryptocurrencies. It contributes to the emerging literature on fundamentals driving cryptocurrency prices by identifying "Natural Disasters"" as another major catalyst of the hash rate other than the energy and mining devices prices. The dominance of Chinese miners in the cryptocurrency mining business builds an imperfect competition in the mining industry. The presence of an imperfect competition leads to a positive correlation between hash rate (or cost of mining) and price of BTC. Therefore
Nature as a Defense from Disasters: Natural Capital and Municipal Bond Yields
I examine the value of climate change mitigation strategies such as nature conservation in municipal bond markets. I find that the market starts to price the value of natural capital after an extreme weather event. Natural capital protection could decrease the county's cost of debt by as much as \$1 million for an average bond. Bonds tied to specific infrastructure projects experience a larger yield increase than general-purpose bonds. The effects of mitigation strategies impact the county with the natural capital and its neighbors. More broadly, I find that natural capital loss is related to population migration and a decrease in personal income, with counties dependent on farming suffering the most. Overall, this paper shows that financial markets price the value of mitigation and highlights the critical role of nature as a shield from natural disasters.
Digital-Money Never Sleeps: New digital-currency indices
In this poster, I introduce three new families of indices around the digital-currency space: on cryptocurrency uncertainty (UCRY Indices), on cryptocurrency environmental (ICEA), and on central bank digital currencies (CBDC indices). These are developed from largescale text mining to reflect the mimetic and emergent nature of the issues. We show that these indices capture well movements and moments in the key digital-currency space. We further apply the IRF and the FEVD to analyse the structure shocks of these indices on other financial and economic variables. Moreover, we decompose these indices’ historical evolution into various drivers. In addition, we apply DCC-GJR-GARCH model to investigate interconnections between CBDC indices and financial variables.
Occupied Investors: The Effect of Foreign Military Presence on Local Investors Asset Allocation
Does foreign culture influence economic decisions of individuals in their native environment? This paper provides evidence on a horizontal cultural transmission channel affecting financial decision-making of individuals. Extending the view on cross-cultural transmission I show that cultural exposure does affect individuals in their own (native) cultural environment when exposed to foreign influences. German retail investor portfolio holdings show a sustained effect of historic exposure to U.S. military presence in the investor’s local environment. The negative perception of large-scale foreign military presence translates into lower likelihood and levels of participation in the U.S. stock market and other foreign stock markets, exacerbating investor’s home bias. Avoidance of U.S. (foreign) equity markets leads to less diversified portfolios due to more concentrated stock holdings. Affected investors’ portfolios also show lower mutual fund shares further reducing overall diversification. These results are robust to socio-demographic and economic controls. Even after controlling for post-war occupation zones in Germany which show opposing effects for cultural transmission from general American presence the results are still in place.
Opioid Epidemic and Mortgage Default
This paper studies the impact of the opioid epidemic on households' mortgage defaults in the United States. Controlling borrower characteristics at mortgage origination, I find that higher opioid overdose mortality rates are associated with higher delinquency rates among prime borrowers. I establish a causal link between the opioid crisis and mortgage defaults by exploiting the variation in oxycodone supply. I document that the result is more pronounced in areas with more insurance coverage rates, suggesting the role of supply factors in driving the epidemic. Income or unemployment growth rates are not explanatory mechanisms for this effect. In contrast, I present evidence that depressed local house prices due to the epidemic have caused more defaults through home equity channel. Accordingly, I show that lenders approve fewer loans to opioid-afflicted areas, indicating pronounced search frictions for the constrained households living in these neighborhoods. Overall, my findings contribute to the current discussions on the direct and indirect costs of the opioid epidemic.
Partisanship in Mutual Fund Portfolios: Biased Expectations or In-Group Favoritism?
Partisan bias in fund portfolios is the effect of fund manager's political affiliation on portfolio allocation decisions. I study two potential channels of this bias: biased expectations where managers become optimistic (pessimistic) when their party comes in (goes out of) the government, and in-group favoritism where managers have higher holdings of politically aligned firms. I find strong evidence for the biased expectations channel. Managers misaligned with the incumbent party underweight value, small, and volatile stocks, and overweight momentum stocks. However, contrary to past literature, I find no evidence for in-group favoritism. I also document a partisan bias in holdings of stocks exposed to politicized topics (COVID-19 and Brexit) but limited evidence for past pandemics (H1N1, Ebola and Zika).
Peer Effects in Financial Expectations
I provide causal evidence that neighborhood financial expectations affect individual financial expectations. I instrument for neighborhood financial expectations with average financial expectations of neighbors' nonlocal family members. Consistent with social interaction driving this effect, I show that social individuals are more influenced by neighborhood financial expectations. Additionally, I provide evidence that individuals who expect their financial situation to improve are less likely to save. This suggests that surveyed expectations reflect actual expectations and that individuals act in accordance with their expectations. Finally, I show that individuals who take neighborhood expectations into account form more accurate expectations.
This paper studies how political agency affects financial markets. Policymakers aim to enact their preferred policies to minimize carbon emissions or maximize output subject to political constraints. When governments and voters disagree over the optimal policy, policymakers endogenously choose opaque policies. By making the learning problem harder for voters, governments can delay or avoid electoral discipline. Greater policy opacity concurrently increases investor uncertainty over future cash flows. I show empirically that these dynamics have tangible effects on asset markets. Option-derived proxies for policy uncertainty and stock price volatility are differentially elevated after environmental policy announcements by governments with preferences different from that of a voting majority of their constituents.
Political Connection and Corporate Litigation: Evidence from a Quasi-Natural Experiment
This paper studies the causal impacts of political connection on corporate litigation. Specifically, by exploiting the enforcement of an unanticipated depoliticization regulation—the Chinese Communist Party's Rule 18 which forces politically connected directors to resign from public firms, we investigate how acquired favoritism and pro tection from political connection impair the effectiveness and fairness of the judicial system. We show that the weakening of political connection results in higher likelihood of and larger monetary amounts involved in lawsuits against connected firms. The effects are more pronounced for non-state-owned enterprises, financially distressed firms, and firms in regions with weak legal institutions. Furthermore, we find that cases with high information asymmetry largely drive the effects. Overall, our pa-per demonstrates that constraining board political connection can mitigate the bias in corporate litigation and level the playing field for litigants.
Product Life-Cycle and Initial Public Offerings
The paper examines how firms' product life-cycle (PLC) influences their trade-off between benefits and costs of going public. We construct the PLC measure by performing a textual analysis on S-1 registration statement for initial public offering (IPO). We show that firms with a more product-innovative PLC are more likely to complete the IPO even though they face higher underpricing and offer a lower fraction of equity at IPO. These firms conduct less seasoned equity offerings, payout fewer dividends, and conduct fewer acquisitions after IPO. The findings demonstrate that firms with diverging PLC differently weigh raising capital through IPO, information asymmetry with investors, and revealing information to competitors. To establish causality, we use an instrumental variable approach with the average PLC of similar public firms as the instrument for an IPO firm's PLC as well as a difference-in-differences approach. Our paper offers novel evidence on a previously under-explored economic force regarding going public: firms' product life-cycle.
Quantile Approach to Asset Pricing Models
This paper develops a generalization of the Hansen-Jagannathan bound that incorporates information beyond the mean and variance of returns. The resulting bound compares the physical and risk neutral distribu- tion for every τ -quantile, where τ ∈ (0, 1). An empirical application with S&P500 return data shows that the new bound is stronger than the Hansen-Jagannathan bound for small values of τ. The long run risk model cannot reconcile this feature of the data, due to the absence of disaster risk. I extend this finding using conditioning information and document that disaster risk is time-varying, using a semiparametric approach. I also propose a new measure of quantile forecastability and show that many stylized facts about the equity premium carry over to the quantile setting.
Bank Ownership and Product Market Competition
This paper studies how bank ownership of industrial firms affects their market power. We find that bank ownership increases firms' markups, while bank ownership of industry rivals reduces firms' markups. Using bank mergers that generate exogenous shocks to bank ownership of industry rivals, we employ a difference-in-differences analysis to establish causality inference. The mechanism analyses show that the decreased markup effect is stronger for competitive industries and R&D intensive firms. Besides, firms are more likely to switch banks, especially when banks have more private information of them. We also find increased costs of loans for the affected firms.
Real-Time Predictability of Mutual Fund Performance Predictors
Researchers have discovered abundant evidence that mutual fund performance is predictable in the cross-section ex post. This paper studies the ex ante predictability of 12 well-known predictors for fund performance from investors’ perspective. Exploiting two types of fund picking strategies with either rule-based approach or machine learning methods, I find that utilizing machine learning can deliver superior real-time economic gains for investors with fund short-term performance being the primary driver underlying predictability. Moreover, using a novel approach to decomposing fund performance, I discover that investors’ flow response to predictor-implied performance exhibits strong variations across predictors. These results suggest ex ante predictability as the compensation for employing costly algorithms to search for skilled managers.
Ruling with Ideology: Politician Belief and the Decisions to Privatize
This paper identifies politician ideology as a fundamental impetus for privatization. Focusing on the world's largest privatization wave in China around 2000, I investigate how provincial leaders' beliefs on the relative merits of state and market shape the substantial regional variation in privatization intensity. a one-standard-deviation increase in a provincial governor’s (party secretary’s) communism score is associated with 1.2% (1.1%, albeit statistically insignificant) fewer SOEs being privatized. Moreover, governor ideology affects privatization through both provincial owned SOEs and subordinate governments owned SOEs, whereas the secretary ideology operates only through the latter. Further analyses indicate that firms privatized by pro-communism governors (but not pro-communism secretaries) also achieve lower post-selling efficiencies. My findings suggest that individual ideology can triumph over the authoritarian institutional norm and manifest itself in high-stake decisions. I also pinpoint politician preference as a new class of determinants for privatization.
Secret Behind Zeros: Round Number Bias in Consumer Lending
This paper provides a unique setting that disentangles the choice dominated by behavioral bias from that dominated by financial constraint. By studying the largest online consumer lending platform, I provide the first empirical evidence that loan amount choice, round versus non-round, contains borrowers’ unobservable private information about their future creditworthiness and ability to repay. Controlling for all borrower characteristics recorded at loan origination, I robustly find that individuals who choose non-round-number loans are 2 percentage points more likely to default than those who choose round-number loans and unintentionally take the arbitrage in this inefficient lending market. Further examination shows that institutional investors largely mitigate this extra default risk through their screening process, and the cost of this extra default risk is transferred mostly to retail investors.
Sequential Learning, Asset Allocation, and Bitcoin Returns
A new class of asset often comes with unprecedented uncertainty. For optimal asset allocation, rational investors must learn about the joint dynamics of new and existing assets. Bitcoin's digital gold narrative provides a unique laboratory to test such a hypothesis. We find that an increase in investors' estimate on correlation between Bitcoin and the US stock markets strongly predicts lower Bitcoin returns next day. The same empirical pattern appears in out-of sample predictions, global equity markets, and other cryptocurrencies. Our stylized static model quantitatively explains the return predictability pattern in light of asset allocation practices and investors' learning on time-varying correlations.
Serial Dependence in the Stock Market: What Can We Learn from Derivatives?
This paper presents a Q-approach to study the serial dependence in the stock market using the derivatives market information only. In theory, the conditional moments of spot and future returns quantities, under the real-world probability measure, can be spanned by no-arbitrage prices of contingent claims that involve index options, VIX futures, and VIX options. As a result, we obtain analytical formulae of market autocorrelation and regression coefficients on returns over two consecutive periods. The method is free of distributional assumptions, robust to different choices of pricing kernel process, and provides a real-time conditional point of view on the stock market. From a forward-looking perspective, we consistently document an upward market trend in the long run but with moderate reversal in the short term. Moreover, we demonstrate that the market return dynamics inferred from derivatives is comparable to that from “learning” the historical stock market prices with fading memory, and the short-term reversal identified by the derivative market is economically relevant.
Share Repurchases: Riding on the Wave of Uncertainty
We document that uncertainty contributes to waves of share repurchases, where monetary policy uncertainty plays a central role in influencing payout policy. In times of high uncertainty about future financial conditions, firms have a precautionary demand for cash and hence pursue a more conservative payout policy by reducing share buybacks. Empirically, we find that this is reflected in leverage and credit spread factors that negatively impact the likelihood of share buybacks. For our sample of buyback transactions in the Economic and Monetary Union of the European Union between 2000 and 2020, the observed cyclicality of the buyback likelihood is particularly driven by variation in prevailing liquidity conditions and uncertainty about monetary policy. This relationship is even more pronounced in the post-quantitative easing period (2010-2020), as expansionary monetary stimulus appears to have made monetary policy an important source of uncertainty for a firm's repurchase decision.
Rise of Superstar Firms and Fall of the Price Mechanism
Since the 1980s, capital allocation efficiency has been deteriorating in the United States. This paper argues that the rise of (superstar) firms and their cash hoarding behavior are reasons. I introduce entrepreneur-manager assignment and corporate risk management into a standard continuous-time heterogeneous agent model with incomplete markets. In this way, Coase (1937)’s firm-(financial) market boundary exists in general equilibrium, and the price mechanism is bounded by corporate internal financing as there is no market to equalize the marginal value of internal resources across firms. Therefore, self-financing (through safe assets) increases misallocation. The scale-related technical change in the 1980s increases the earnings-quality gradient sharply in the right tail, which not only generates a winners-take-most phenomenon but also makes current winners inherently riskier and rely less on external financing. This risk redistribution nature of technical change expands the internal financing region and impairs the capital allocation efficiency. When taken to the data, the model can quantitatively match some important macro-finance trends, and it shows that the area disciplined by the market system has declined by about 11% during the past forty years.
This study examines analyst “silence”: a previously unexplored tendency of research analysts to suddenly stop publishing research on a covered stock. An analyst may go “silent” and withhold information from clients if they have a private motive for valuable information that they have collected. When an affiliated asset manager purchases a stock that a (sell-side) analyst goes silent on, the stock displays annualized abnormal returns of 12-13%. The long-short trading strategy arising from this phenomenon produces tradeable profits, in contrast to the “silence” of analysts unaffiliated to any asset managers. The prevalence of strategic analyst silence increases with stock volatility, reducing the usefulness of analyst-produced information for their clients and raising welfare and market efficiency concerns. These patterns highlight a previously unrecognized conflict between analysts and their clients which may be “silently” harming the information environment.
Skills and Sentiment in Sustainable Investing
We document a significant difference in the returns of sustainable investing across investor types. Investors with strict ESG mandates earn 3.1% less than flexible investors. The mechanism is that flexible investors are able to react on expected ESG improvements. They buy stocks that subsequently experience ESG score increases. After ESG improvements have realized, demand from strict mandate investors pushes up stock prices, resulting in positive returns for flexible investors. These returns are higher when accompanied by rising climate sentiment, as seen during the 2010s. Our channel accounts for 51% of the return difference between strict and flexible ESG investment mandates.
Social Investing and Hype in the Stock Market
Online group discussions about stocks intensify as trading becomes easier to access and messaging technology develops. Using data from investing-related chat rooms, I find that live group chats help investors find high alpha stocks than individual posts and comments in investing forums. Moreover, sentiment is less important than identifying which stocks are being discussed. I also create a hype measure that positively predicts trading volume, stock volatility, and future returns. Hyped stocks tilt towards small-cap and growth stocks. I find that higher returns are driven more by the continuity of hype than by the day a stock becomes hyped. The returns show an upward drift, and insiders are less likely to sell after they become hyped, suggesting that the trades are informed.
Social Proximity to Start-Up Funding
This paper examines whether aggregate social networks influence start-up firms’ funding characteristics. We find that start-ups in U.S. counties with higher social proximity to start-up funding (SPF) attract more capital and investors than their lower counterparts. Using historical travel costs between counties as an instrument, we show that this relation is likely causal. We also find that the strong, positive relation between SPF and funding characteristics holds for minority-founded (female or black) start-ups. Consequently, start-ups in regions with higher SPF exit faster through acquisition than lower counterparts. Our results highlight the importance of aggregate social connection for early-stage funding.
Stock Returns in Global Value Chains: The Role of Upstreamness and Downstreamness
This paper studies how upstreamness and downstreamness affect stock returns in global value chains. Up- and downstreamness measure the average distance from final consumption and primary inputs, respectively, and are computed from world input-output tables. We show that downstreamness is a key driver of expected returns around the globe, whereas upstreamness is not. Firms that are farthest away from primary inputs earn approximately 5% higher returns per year than firms that are closest. The effect is found within and across countries and suggests that investors perceive supplier dependence in global value chains as an important source of risk.
Subjective Learning of Trading Talent: Theory and Evidence from Individual Investors in the U.S.
Recent studies show evidence that investors learn about their trading abilities. This paper focuses on understanding how investors learn about their talent and proposes a unifying framework that explains many puzzling facts about individual equity investors. In my model, the investor forms subjective beliefs about both the expected return of the current stock-in-holding and her trading talent represented by the expected return of the next replacement stock, and updates beliefs through learning with fading memory. I calibrate the memory decay parameters to individual trading records, and show that talent learning is about 7 times more sensitive to return signals than stock-in-holding learning. Consequently, the model indicates that stock switching always happens following good performance of the current stock because switching requires a sufficiently large wedge between expected returns of the replacement stock and the current stock to cover the fixed cost, which strongly predicts the timing of investors' buying in a learning perspective. This framework also accounts for the performance-contingent trading intensity and attrition, and explains why a negative shock would lead to attrition when an investor has several years of experience, which is inconsistent with the decreasing-gain updating under standard Bayesian learning.
Target Information Asymmetry and Post-Takeover Performance
This paper examines the impact of target information asymmetry (IA) on US acquiring firm’s post-takeover performance over the period 1990 to 2015. Prior theoretical research presents a contradictory impact of target IA on post-takeover performance, which either poses threats to acquiring firms due to an adverse selection problem or gives rise to superior performance by obtaining private information. Our results support the private information theory. We also report a stronger relationship for more innovative deals, especially when the target has high R&D intensity. We also show that stock financing for these deals provides additional improvement in post-takeover performance, consistent with possible ‘championing culture’ benefits and with stock mitigating part of the increased risk for more innovative deals. We provide some evidence to support that private information obtained relates to pre-takeover innovation, and show that acquirers significantly increase R&D investment post-takeover for deals financed with stock. We employ methods to address possible econometric concerns with selection and omitted variable bias.
The Core, the Periphery, and the Disaster: Corporate-Sovereign Nexus in COVID-19 Times
We study how the COVID-19 pandemic reshaped the relation between corporate and sovereign credit risk in the cross-section of countries in the European Union. Surprisingly, the outbreak triggered higher elasticity of corporate to sovereign CDS spreads in core countries, which realigned to that of peripheral countries, with lower fiscal capacity, for which the impact of the pandemic on the elasticity was essentially muted. During the pandemic, we observe systematic departures of actual CDS from those implied by a standard structural model of default for larger firms in core EU countries with budgetary slackness. We interpret this evidence in light of a disaster-risk asset pricing model with bailout guarantees and defaultable public debt. Based on the model and a synthetic control method, we show that CDS-implied risk-adjusted bailout guarantees over the medium term were about three times larger in the Core than in the Periphery.
The Deposits Channel of Aggregate Fluctuations
This paper presents a new mechanism through which the geography of bank deposits increases financial fragility. We document the within-bank geographic concentration of deposits -- 30% of bank deposits are concentrated in a single county. We combine this within-bank geographic concentration of deposits with local natural disaster-induced property damages to construct novel bank deposit shocks. On aggregate, these shocks can explain 3.30% of variation in economic growth. Local disaster shocks result in aggregate fluctuations through their effect on deposits, which negatively affect bank lending. Financial frictions such as regulatory constraints, informational advantages, and borrower constraints are critical for the aggregation of shocks.
The Diminishing Impact of Monetary Policy on Asset Prices Around Non-FOMC Macroeconomic Announcements
I examine the effects of monetary policy surprises on asset prices around non-FOMC macroeconomic announcements that are directly relevant to the Fed's monetary policy decisions. While FOMC announcements are known to have similar effects during periods of conventional and unconventional monetary policies, I show that non-FOMC announcements affect asset prices much less in the latter period. Moreover, bond premium, volatility and the overall resolution of uncertainty decrease on these announcements. These findings are described in an information framework. Taken together, the evidence suggests unconventional monetary policies deter market's ability to anticipate Fed actions, which has implications for its transmission to asset prices.
The Distress Puzzle and Credit Forbearance
Using a unique data set on credit forbearance agreements, I provide evidence that the reduction in firm risk following execution of a credit forbearance agreement contributes, in part, to the well-documented distress anomaly. These findings are consistent with prior literature hypothesizing that post-default shareholder bargaining power partially explains the distress anomaly. Distressed firms experience a decline in returns and market beta following entrance into a forbearance agreement. A zero-investment trading strategy that first sorts firms by financial distress and then by entrance into a forbearance agreement earns statistically and economically significant six-factor alpha of up to 3.52% per month. The model’s performance is stronger for firms with recent credit forbearance agreements.
The Downstream Channel of Financial Constraints and the Amplification of Aggregate Downturns
We identify a novel channel through which financial constraints propagate in the production chain. Firms experience greater valuation losses during industry downturns when their suppliers are financially constrained. Exploiting recent developments on production network data of all listed US firms, we link firms vertically and find that downturn effects are stronger when: (i) suppliers are more constrained; (ii) firms depend more on specific inputs; and (iii) suppliers are more concentrated. The effects are attenuated or muted when suppliers manage to keep high levels of accounts receivables, suggesting trade credit as a mechanism through which the downstream channel operates. Our findings uncover two network implications of financing constraints: stronger contagion of negative shocks through supplier-customer links and the amplification of customer industries' aggregate valuation losses. Our results lend support to policies that facilitate trade credit in upstream segments during crises.
The Effect of Local Market Concentration on Deposit Price Dispersion
I examine the effect of the local market’s bank concentration on the price dispersion of the deposit products. By using the Interstate Branching Deregulation status of a region as the instrumental variable for the bank concentration, I show that the local market’s bank concentration has a negative effect on the price dispersion of the deposit products. This effect is not monotonic and holds only for the core deposit products. I also find that price dispersion increases in less concentrated market because of branch entry as the entrant banks always offer higher prices over the incumbent banks. The effect exists during loosening or tightening period of the monetary policy but goes away during the upper zero bound of the Fed Rate.
The Effect of Labor Unions on Municipal Bonds
We present three findings on the effects of local unions on the municipal bond market. First, municipal bond issuers in areas with higher public-sector union density have higher issuance costs. Private-sector unions only affect issuance costs when an issuer is exposed to strong private-sector union power. Second, by employing a regression discontinuity design using local variation in the vote share of union elections, we find that closely won union elections lead to significantly higher yields on the secondary market. Third, state-level legislation that restricts the collective bargaining power and the possibility of a strike affects the issuance costs of municipal bonds. Our findings suggest that union density is viewed as a risk factor by municipal bond investors.
The Effects of Capital and Liquidity Requirements in a Macroeconomic Dynamic Model of Banking
This paper studies the quantitative impacts of Basel-style capital and liquidity requirements on bank lending, bank liquidity holdings and interbank trading activities. We develop a model in which banks are subject to business cycle variations, are financed by deposits and equity, and transform these liabilities into loans, liquid assets, and interbank lending. Banks are exposed to systematic credit and liquidity shocks and idiosyncratic liquidity and credit profit shocks, where the idiosyncratic shocks can be coped with through the interbank market. We find that (1) banks’ liquidity is countercyclical and liquidity requirements are then more effective in mitigating banks’ liquidity issues in economic expansions, (2) the benefits of liquidity requirements are at the cost of lowered social welfare, and (3) there is a U-shaped relationship between interbank trading volume and the liquidity requirements, and the recent liquidity required at 100% (for LCR and NSFR) seems too strict to limit banks’ excessive reliance on the interbank market. Liquidity requirements (both for LCR and NSFR) around 65% are the
The Effects of Information Acquisition in M&As: Evidence from SEC EDGAR Web Traffic
This paper studies the effects of information acquisition in mergers and acquisitions (M&As). Information acquisition, proxied by downloads of filings on the SEC EDGAR website, improves the market’s assessment of deal synergies. Specifically, the information acquisition about merging firms, industry rivals, and supply-chain firms enhances the relation between combined announcement return and post-merger performance in merged firms. The informational role is more important for mergers with greater institutional downloads and more intensive institutional trading activities. Merging firms’ stock prices react more to new information about the merger. Further, information acquisition in merging firms improves market informativeness about both production synergies and financial synergies achieved by the merger. Overall, this paper provides supportive evidence that information acquisition activities improve the efficiency of market valuation in mergers.
The Evolution of Market Efficiency Over the Past Century
I combine a hand-collected sample of earnings announcements from the Wall Street Journal over the years, 1934-1971, with more recent data from Compustat, and document a striking U-shaped pattern in the evolution of market efficiency over the extended period, 1934-2018. In terms of investors’ response to both firm-specific and market-wide news, markets are more efficient during the early and late years in this extended sample, while they become less efficient in the middle decades. I argue that this U-shaped pattern in the degree of market efficiency over time has been driven by two distinct economic dynamics. While the recent evolution in information-processing technology has led to more efficient markets in the later periods, the surprisingly high degree of market efficiency in the 1930s and 1940s reflects the greater relative importance of earnings announcements as a critical source of information that commanded investor attention, at a time when there was less overall information to process and fewer alternative information venues to consider. Overall, these results highlight that the evolution of market efficiency has not followed a linear path, but rather, divergent economic forces have caused the U-shaped pattern in market efficiency over time.
The Industry Expertise Channel of Mortgage Lending
This paper documents an industry expertise channel that reduces the information asymmetry between banks and mortgage borrowers. This channel is a result of information spillover from a bank’s specialization in corporate lending to its mortgage lending. We find that banks allocate more mortgage credits to counties with shared industry specialization, especially when the information asymmetry or borrower risk is high. Further tests show that mortgages originated through the channel contain more soft information and have better performance. The findings suggest that information from the channel improves banks’ screening and monitoring efficiency in the mortgage market.
The Information Content of Trump Tweets and the Currency Market
Using textual analysis, we identify the set of Trump tweets that contain information on macroeconomic policy, trade, or exchange rate content. We find that informative Trump tweets reduce speculative trading in foreign exchange markets, with a corresponding decline in trading volume, volatility, bid-ask spreads, and induce a positive bias in returns reflecting Trump’s (optimistic) views on the U.S. economy. Two-thirds of his informative tweets are optimistic. This bias serves as a diversion strategy from negative media coverage. We rationalize these results within a model of Trump tweets revealing economic content as a public signal that reduces disagreement among speculators.
The Pricing of Continuous and Discontinuous Factor Risks
This study considers a continuous-time version of the Fama-French (2015) five-factor model, explicitly allowing stocks' exposures on the factors' continuous, jump, and overnight movements to be different. Our results show that stocks' continuous, jump, and overnight betas with respect to a given factor can be very different and are only weakly related. We find strong evidence for a positive pricing of continuous market exposure and a negative pricing of overnight market exposure whereas jump market exposure is not priced. This finding contradicts prior empirical evidence indicating a positive pricing of jump and overnight market exposures but zero pricing of continuous market exposure. Moreover, exposures to the size, value, profitability, and investment factors' continuous risks are mostly negatively priced while exposures to their overnight risks are positively priced, suggesting that these factors' return premia are compensation for exposure to the factors' overnight risks. Jump exposures are in general not significantly priced.
Real Effect of Competition Laws: International Evidence
We employ the statutory laws that regulate competitions among firms to examine the effect of competition on firms’ financing and investment decisions around the world. Using a large sample of 206,713 firm-year observations from 59 countries, we find evidence that the stringency of competition laws spurs improvement in governance that positively influences firms’ access to external financing and investment. This finding is robust to several sensitivity tests, including alternative fixed effects, alternative sample compositions and a first-difference change analysis. A breakdown of the competition law index into its two subcategories, namely, authority and substance, reveals that the effect is mainly driven by the authority subcomponent of the competition law index. In cross-sectional analysis, we find that the positive association between the stringency of competition laws and external financing and investment is stronger for firms from countries with weaker institutional environments. Using a sample that covers diverse industries and many countries, our study improves the understanding of the effects of market competition, particularly, how variation in institutional settings matters for the success of competition laws and should be of value to policymakers.
The Role of Financial Expert CEOs in Mergers & Acquisitions
Does financial work experience help CEOs make decisions on Mergers & Acquisitions (M&As)? Using a sample of CEOs from S&P 1500 firms from 1992-2018, we find that financial experts underperform in takeovers. CEOs with financial work experience are bad bargainers and create fewer synergies with targets. However, they seem to understand the value-destruction accompanying takeovers and thus engage in fewer deals. We further suggest that financial expertise comes at the expense of having expertise in other dimensions. When CEOs gain industry expertise, their financial expertise is the icing on the cake. Meanwhile, financial expert CEOs disproportionately prefer public targets, which prove to be generally associated with value destruction.
The Role of Stock Indices in Analyst Career Outcomes
Random changes in firms' stock index membership affect sell-side analysts' career outcomes. We study the role of firms' movements between Russell 1000 and 2000 indices that cause discontinuous changes in institutional ownership around the index threshold and hence in the importance of analysts covering these stocks. Firms moving from the bottom of Russell 1000 to the top of Russell 2000 significantly increase an analyst's likelihood to move to a higher-status broker. This beneficial outcome for the analyst is reflected in analyst recommendations. For firms that are just above the index threshold (i.e., that might move to Russell 2000 if their share price decreases slightly), analyst recommendations are significantly more negative in April, the time of defining the index weights that determine index membership.
The Tradeoff between Discrete Pricing and Discrete Quantities: Evidence from U.S.-listed Firms
Economists commonly assume that price and quantity are continuous variables, while in reality both are discrete variables. As U.S. regulation mandates a one-cent minimum tick size and a 100-share minimum lot size, we predict that less volatile stocks and more active stocks should choose higher prices to make pricing more continuous and quantity more discrete. Despite heterogeneous optimal prices, all firms achieve their optimal prices when their bid–ask spreads equal two ticks, when frictions from discrete pricing equal those from discrete lots. Empirically, our theoretical model explains 57% of cross-sectional variations in stock prices and 81% of cross-sectional variations in stock liquidity. We find that most stock splits move the bid–ask spread closer to two ticks and that correct splits contribute 94 bps to split announcement returns. Optimal pricing can increase median U.S. stock value by 106 bps and total U.S. market capitalization by $93.7 billion.
Three Aspects of Green Bonds
This paper examines three fundamental questions regarding green bonds – ‘how stockholders react to green bonds issuance in different countries? ‘which firms issue green bonds?’, and ‘who supports their issuance?’ The stylized facts suggest that green bonds issuance is highly concentrated among few firms (and their subsidiaries) in the US and Europe but diversified in the Asian region. On analyzing the stockholder’s reaction to green bonds issuance in 19 countries, I find that there is a difference in market reaction (in magnitude and in direction) to the issuance of green bonds in all these countries. I also find that firms with low environment score, low ESG score, high unscaled carbon emissions, and with no target emissions issue more green bonds compare to others. The latter result supports the signaling hypothesis. Also, only domestic (and not foreign) institutional investors support the issuance of the green bonds which implies that there is a home-bias effect. Similar to market reaction, the latter result also varies across the countries. In sum, this paper suggests that it is crucial to understand the intricacies in the corporate green bond issuance to correctly emulate stockholders' reaction, to highlight the identity of green bond issuers, and to know what kind of institutional investors supports the green bonds issuance.
Trust as an Entry Barrier: Evidence from FinTech Adoption
This paper studies the role of trust in incumbent lenders (banks) as an entry barrier to emerging FinTech lenders in credit markets. The empirical setting exploits the outbreak of the Wells Fargo scandal as a negative shock to borrowers' trust in banks. Using a difference-in-differences framework, I find that increased exposure to the Wells Fargo scandal leads to an increase in the probability of borrowers using FinTech as mortgage originators. Utilizing political affiliation to proxy for the magnitude of trust erosion in banks in a triple-differences specification, I find that, conditional on the same exposure to the scandal, a county experiencing a greater erosion of trust has a larger increase in FinTech share relative to a county experiencing less of an erosion of trust. Estimating treatment effect heterogeneity using generic machine learning inference suggests that borrowers with the greatest decrease in trust in banks and the greatest increase in FinTech adoption have similar characteristics.
Unconventional Monetary Policy and Household Credit Inequality
Does unconventional monetary policy have a distributional effect on household credit? To answer this question, I use granular data from the European Central Bank’s Household Finance and Consumption Survey covering 17 countries in the euro area and compare household credit in the pre–Asset Purchase Programme (APP) period with household credit in the post APP period. I find that the credit gap between the top and bottom of the distribution widens, and the largest increase in credit after the policy implementation is among middle-wealth households. Recentered influence function regression and decomposition results suggest two potential policy transmission channels for household credit inequality: (1) the credit risk channel, which increases inequality through the assets valuation effect, and (2) the credit constraint channel, which reduces inequality by facilitating access to credit for low- and middle-wealth households. Finally, an investigation of household asset portfolios finds that property ownership and rising housing prices were the key drivers of household financing decisions after the APP went into effect.
US Wealth Shares, the Dollar and Global Risk Premia
I uncover a novel stylised fact: the US wealth share is countercyclical w.r.t the global economy. These wealth share dynamics are tightly connected with the dollar: rises in the US wealth share coincide with dollar appreciations against the ROW. These facts challenge exorbitant privilege (EP) theory and their risk sharing view of these international dynamics. I argue instead that a global risk premium story can account for these patterns. This argument is formalised using a frictionless two country, two-good model with recursive preferences and heterogenous global shock exposures. Thus risk premia, not risk sharing, is the key economic force driving the joint dynamics between the US wealth share, the dollar and the global economy.
Wall Street Goes Dark: Venue Selection During the COVID-19 Market Crash
We investigate the venue choice of traders and price discovery during the COVID-19 market
crash, using a sample of 801 stocks in the U.S stock market. We classify trading venues based on
their degree of execution immediacy as: lit venues, continuous dark pools, and scheduled dark
pools. We find that the market share of dark pools decreases significantly in the first week of the crash, before increasing in subsequent weeks. The decrease in the market share of dark pools in the first week of the crash is more pronounced for scheduled dark pools, which are expected to
provide the lowest degree of execution immediacy. We further classify firms based on return
volatility during the crash, and document a negative relationship between return volatility and
dark trading. Our results also show that price discovery during the crash is weaker (better)
when the historical liquidity in buyside (internalization) pools is higher.
When Green Meets Green
We investigate whether and how the environmental consciousness (greenness for short) of firms and banks is reflected in the pricing of bank credit. Using a large international sample of syndicated loans over the period 2011-2019, we find that firms' are indeed rewarded for being green in the form of cheaper loans---however, only when borrowing from a green consortium of lenders, and only after the ratification of the Paris Agreement in 2015. Thus, we find that environmental attitudes matter "when green meets green"". We further construct a simple stylized theoretical model to show that the green-meets-green pattern emerges in equilibrium as the result of the third-degree price discrimination with regard to firms' greenness."
Why Do CEO Compensation Schemes Feature Convexity? Evidence from a Natural Experiment
We provide causal evidence of CEO compensation schemes featuring convexity to provide risk-taking incentives. Specifically, we leverage the Federal Trademark Dilution Act signed in 1996 which granted additional legal protection to selected trademarks against dilution. We argue that this made risky product-market expansion more appealing to shareholders of firms with protected trademarks because product differentiation is guaranteed. We show that firms significantly increase the convexity of CEO compensation in response to exogenous increases in investment opportunities. And this increase in convexity is more pronounced for firms whose brands are well recognized, products are more substitutable, and CEOs have more career concerns.
Why Do Innovative Firms Sell Patents? An Empirical Analysis of the Causes and Consequences of Patent Transactions
In this paper, I analyze the secondary market transactions of patents from public assignor (i.e., seller) to assignee (i.e., buyer) firms. In particular, I study the causes and consequences of public assignor firms selling some of their patents. I document that firms with higher innovation productivity or innovation quality but with lower production efficiency are more likely to sell patents distant from their operations. Further, patents with lower economic value but higher scientific value are more likely to be sold. In terms of the consequences of patent transactions, I document that in the three years after patent transactions, assignor firms on average experience a positive and statistically significant improvement in their operating performance. In addition, their stocks enjoy a positive and significant long-run buy-and-hold abnormal return (BHAR) following these patent transactions. This pattern is robust to different holding periods and benchmark portfolios against which the long-run buy-and-hold return is calculated. I document one possible underlying mechanism driving these results, which is that assignor firms increase their focus after patent transactions.