« 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.
An International Study of Common Ownership, Control and Competition
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&Vives (2021), I measure intra-industry and inter-industry common ownership and empirically investigate their respective effects on competition. Preliminary results suggest that intra-industry common ownership increases firm markups whereas inter-industry common ownership decreases them, as predicted by Azar&Vives (2021). 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, Enhanced Patent Publication, and Inventor Human Capital Reallocation
How does the enhanced patent publication affect inventor human capital reallocation? Leveraging (i) the American Inventor’s Protection Act that enhances US patent publications as a natural experiment, (ii) the plausibly quasi-random assignment of sluggish patent examiners, and (iii) a large dataset tracking inventors’ career paths, I find that after the AIPA, inventor mobility increases, particularly for inventors who had long historical patent publication delay. Enhanced patent publication promotes the information of inventors who were previously overlooked and whose human capital may have been misallocated. Increased mobility enables inventors to reallocate to better-matched companies and increase their innovation productivity. Suggestive evidence indicates that enhanced patent publication reduces the deterioration of losing inventors and increases the benefit of gaining inventors for innovative firms. The effect is further corroborated by a second strategy exploring staggered Patent Depository Library expansion program that facilitated patent publication in pre-AIPA. This study highlights an important but overlooked aspect of patent publication’s influence on innovation – inventor mobility.
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 relation 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, consistent with a certification
role that the bank plays. This paper adds to the debate on the benefits and
drawbacks of bank’s 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 risk-sensitive capital and penalties when banks have private information about their risk. I find that the Basel framework can be rationalized provided sufficient variation in banks' preferences. I then use hand-collected data on risk model revisions to show that current penalties provide only weak incentives for model improvements and in fact induce misreporting. My model suggests that recent changes in Basel 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.
An increasing amount of assets is managed by benchmark-tracking 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 2017 and implemented over the period from May 2018 to November 2019. This change presents a rare opportunity to estimate the causal impact of discretionary index changes and shed light on cross-sectional implications. We document systematic deviations from the benchmark, in particular for passive funds. Market capitalization, stock liquidity and stock volatility affect how benchmark changes translate to portfolio adjustments. We find that these characteristics moderate the impact of benchmarking changes on financial market outcomes, suggesting that deviations from benchmarking 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 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 for affected (non-financial) firms. 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 empirical results highlight the existence of both channels with the arbitrage channel outweighing the hedging channel.
Climate Change Regulatory Risks and Bank Lending
We investigate how banks adjust their credit supply depending on firms' exposure to regulatory risks related to climate change and banks' own exposure, which originates from their portfolio structure and borrowers’ exposure. Exploiting the Paris Agreement as a shock to banks' perception of regulatory risks, we identify large heterogeneity in credit reallocation depending on whether the borrower stands to gain or to lose from future regulation as well as the region in which the borrower is located, which highlights the importance of the existing regulatory environment. Banks lend relatively more to US firms with a negative exposure to regulatory risks while, in contrast, supplying relatively more credit to European firms that are positively exposed. For US firms there is no differential effect of banks' own exposure whereas we find that the more negatively exposed a bank is, the more it increases its credit supply to negatively exposed European firms.
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.
Community Diversity and Financial Decision Making
This paper demonstrates that community diversity is a key determinant for household-level financial outcomes. Diverse communities are defined as those in which the likelihood of interracial interaction is high. Specifically, individuals residing in racially diverse communities are more likely to invest in public equity markets relative to those residing in racially isolated communities. These findings are robust and cannot be explained by differences in individual characteristics such as income, education and wealth. Evidence suggests that diversity increases confidence in locally acquired information, lowering the subjective cost of stock market participation. Overall, this paper demonstrates that racial diversity improves the financial outcomes for both majority and minority groups.
Competition and 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 analyses how creditor control rights influence the pricing of corporate loans
in a novel dataset of covenant violations combined with borrower, lender, and loan characteristics.
We exploit the discontinuous shift in control rights to creditors after a covenant
violation to isolate this influence. Applying a quasi-regression discontinuity design, we
find that creditors use their control rights to increase loan spreads by, on average, 19% to
47%. The increase in loan spreads is larger for borrowers with a weaker bargaining position
and less bargaining frictions in the lending syndicate. Our data structure also allows
us to support our conjecture by applying additional difference-in-differences and fixed effects
estimations. These tests suggest that the results are not confounded by borroweror
lender-specific unobserved heterogeneity. Our results highlight an important channel
through which lenders influence debt pricing to extract rents from borrowers.
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 loans and public bonds when raising new debt capital. We find that cross-ownership has a significantly negative effect on the firm’s usage of bank loans. Consistent with the channel where cross-ownership improves shareholder governance by internalizing governance externalities and hence reduce the reliance on creditor governance, we find this negative relationship is more pronounced for firms with greater governance externalities. Evidence from a quasi-natural experiment based on financial institution mergers mitigates concerns of reverse causality. Our paper highlight the beneficial role of cross-ownership by reducing the cost of debt financing through internalizing the governance externalities.
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 across the states of India in the 1990s. Exploiting this quasi-random variation in efficiency of debt contract enforcement and using detailed information on product lines produced by manufacturing firms in India, we study the causal effect of debt contract enforcement on product innovation. Our estimates suggest that increase in access to collateral for creditors leads to a significant increase in the product scope of firms. This increase in product scope is confined to firms with initially high tangible assets. These firms also increase sales, exports, profitability, and TFP. Further, they increase investments in plant, property, and equipment, selling & distribution expenditure, R & D expenditure, and imports of raw materials. We also find that these firms become more financially disciplined and increase their repayments.
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 under-capitalized and well-capitalized banks, and also 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?
This paper investigates whether investors overreact to managerial tones in the text of annual reports. Using a machine learning technique on topic analysis, I conduct a decomposition of managerial tones on multiple topics (signals), which are informative about different aspects of firms' value (fundamentals). Focusing on 10-Ks of SP500 constituents, I find the market reacts more to signals which are precise about future fundamentals, indicating that investors extract value-related information from tones while ignoring text uncorrelated with fundamentals. A simple structural model of trading shows that while naive investors overestimate excess returns by 2.75 basis points on average, investors' overreaction pushes up the filing period buy-and-hold excess return by only 0.03 basis points. Cross-sectional analysis shows that overreaction is most salient for high market valuation firms and medium-size ones. The results support the view that investors are not overreacting to managerial tones.
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 competition 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 it 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 to this setting. 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 in terms of forecast error and variance across analysts. I find that the disciplinary effect of competition dominates in the current market, reducing forecast error by 41.40%, at a cost of a 4.22% increase in forecast variance. However, once the optimism incentive is removed, competition increases both forecast error and forecast variance.
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
I employ the seeded Latent Dirichlet Process (sLDA) model in natural language processing to extract the narratives discussed by Shiller (2019) from nearly seven million New York Times articles over the past 150 years. The estimation scheme is designed to avoid any look-ahead bias in constructing the monthly narrative weights. Among the narratives considered, the most important one is Panic, which encompasses various stress- and anxiety-related themes including economic downturns, wars, political tensions, and epidemics. I find that Panic and a narrative index that loads heavily on Panic are strong positive predictors of excess U.S. market return and negative predictors of both realized and implied market volatility. I document empirical support for Panic as a proxy for time-varying risk aversion, consistent with a univariate version of the intertemporal capital asset pricing model (ICAPM). The predictability of narratives over market returns holds at both market and portfolio level and at both monthly and daily interval, and importantly is increasing over time.
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 9.05% and 4.01% 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. EEOC announcements have no impact on stock returns, firm financials, or CEO turnover but lead to human capital risk and workforce reductions 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
Explaining Greenium in a Macro-Finance Integrated Assessment Model
I investigate how firms' environmental responsibilities affect expected stock returns. Using the environmental pillar score from the ASSET4 ESG dataset, I find that greener stocks have lower expected returns. This greenium remains significant after controlling for systematic and idiosyncratic risks. I explain the greenium through event studies showing that green stocks hedge physical climate-change risks. A macro-finance integrated assessment model (MFIAM) featuring time-varying climate damage intensity, recursive preferences, and investment frictions supports the empirical findings. The model implies that climate damages are pro-cyclical, leading to a high discount rate and a relatively low social cost of carbon.
External Market of Workplace Safety
This paper examines if firms boundaries respond to variation in legal risks arising from industrial accidents. If outside contractors are more effective in managing workplace accident-related legal liabilities, then do firms outsource labor-intensive activities to external contractors? Does it affect the overall employment and investment decisions of firms? I address these questions by exploiting an amendment to the Industrial Safety Law in Korea in 2017, which expanded legal liabilities of firms to also include contract workers. I find that affected firms moved away from outside contracting following the amendment to the law. Although reliance on contract work fell, firms did not compensate for it by direct hiring. Consequently, overall employment fell when legal liabilities related to workplace accidents increased. Investments in physical assets fell too. The results are consistent with firms strategically outsourcing riskier jobs to contractors to offload their legal liabilities relating to worker safety. While the law improved visibility on workplace accidents, it adversely affected employment, investment in physical assets, and growth.
Financial Development and Health
In this paper, I examine how development of the financial sector affects health, using a nationwide natural experiment. I exploit a policy of the Reserve Bank of India from 2005 that introduces exogenous variation in bank presence. The objective of the policy is to incentivize banks to set up new branches in underbanked districts, defined as districts that have a population-to-branch ratio above the national average. Using a regression discontinuity design, I compare households in districts that have a ratio just above and just below the national average. I find a strong and robust positive effect on health. Six years after the policy was introduced, households in treatment districts are 36 percent less likely to be affected by an illness in a given month. This positively impacts their economic situation; they gain half a day of work or education and spend significantly less on 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 financial development played an important role. First, households gain access to health insurance. Second, health care providers gain access to credit and improve supply.
Flag Tag: Credit File Disaster Flags As Social Insurance Tags
This paper finds 59.2 million people had a ‘disaster flag’ on their 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 \& 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 event study. I find small average effects of flags on credit scores (1.5-2pp) driven by larger (10-15pp), temporary effects for those with pre-disaster defaults or subprime credit scores. There are no clear effects on credit access (e.g. account openings, credit limits). Finally, the paper evaluates the information value of data masked by flags for lenders’ ability to predict credit risk and compares results to a counterfactual social insurance regime automatically masking all new defaults during natural disasters.
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: Evidence from China
We show that trading suspension, a regulatory rule that temporarily restricts stock
trading, generates stale mutual fund net asset values (NAVs). Using a sample of 3,205
long-lasting suspension events in China between 2004–2018, we find that opportunistic
investors exploit transformed liquidity and allocate capital based on firm-specific news
and fund portfolio holdings: Abnormal flows positively respond to suspended stocks’
unrealized impact on NAVs. Trading suspension generates short-term profits at the
cost of long-term investors, and portfolio disclosure plays a key informational role in
distorting flows. Our findings suggest that trading regulations can have unintended
spillover effects on non-targeted markets.
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. State 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 causes higher productivity and profitability of firms affected by such mandates. First, we show that the effects on performance are more pronounced for firms operating in industries with more expensive labor and lower labor intensity. We interpret these results to suggest that employees who at least marginally prefer sick pay to pecuniary compensation will have higher incentive to exert more efforts, resulting in better firm performance. Interestingly, the latter result is mainly driven by firms headquartered in counties with higher social capital, which are less prone to moral hazard. Second, we confirm that generosity of PSL boosts firm performance by improving employees' health. To this end, we find that the performance improvement is stronger for firms operating in industries with lower share of jobs that can be done at home. Finally, additional tests reveal that increased PSL coverage is associated with higher firm value and more use of leverage. Collectively, our paper demonstrates a Pareto improvement associated with such provision of fringe benefits, and it provides a potential resolution of 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, particularly the effects of making additional predictive signals available to investors. We analyse a panel of systematic traders' investment outcomes, sourced from a FinTech platform that organises trading contests under highly-controlled conditions that allow us to identify learning effects. Investor outcomes improve with experience, and this is also apparent when counterfactually assessing their trading decisions on historical data, suggesting that they make use of historical data to attain their objectives. When additional predictive variables are added to the common part of investors' information sets, investors’ relative performance outcomes improve at higher experience levels. To explain these results, we model an investor as choosing a portfolio by learning from historical data while also taking model uncertainty into account. Empirically, inexperienced investors seem to ignore newly available predictive signals, in keeping with our theory’s predictions.
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 when managers’ actions do not maximize the total value of the firm (Jensen and Meckling, 1976; Myers, 1977; Smith Jr and Warner, 1979). Shareholder-creditor conflicts can induce agency costs in the form of excessive dividends, claim dilution, assets substitution and investments distortions, and lead to a significant wealth transfer from creditors to shareholders and vice versa. Gennaioli and Rossi (2013) show that shareholder-creditor conflict exacerbates around financial distress, while Gilje (2016) empirically demonstrates that shareholder-creditor conflict pushes financially distressed firms to take less risk. Therefore, it is important to empirically understand how the shareholder-creditor conflict of interest affects company value and performance. In this paper, I am going to provide evidence regarding shareholder-creditor conflict using the existence of dual-holders – investors who hold equity and debt claims of the same firm. Similar to Jiang et al. (2010) and Chu (2018) I measure the shareholder-creditor conflict by institutional dual-holding, particularly, dual-holdings of U.S. mutual funds.
Institutional investors own more than two-thirds of U.S. publicly traded equity but they also hold a large fraction of U.S. corporate bonds. The average share of a mutual fund family’s total net assets held in corporate bonds has increased from around 5% in 2008 to 15% in recent years (Gormley and Jha, 2020). Considering the amount of money that institutional investors use and the wide socio-economic impact of their activity, it is crucial to ask what will happen to the firm value and performance if its institutional investors are dual-holders.
I exploit the holdings of U.S. mutual funds as an indicator of shareholder-creditor conflict because mutual funds hold portfolios that consist of diverse stakes in various firms and normally (rarely) the single equity stake does not exceed 10% level on the fund level. Also, mutual funds regularly report their portfolio holdings and voting patterns to authorities. Ultimately, I will be able to answer how the presence of mutual fund dual-holdings affects public companies. Furthermore, I will elaborate on the impact of an increase in dual-ownership stakes over company financial results and will attempt to shed a light on whether dual-holdings can alleviate corporate financial difficulties.
Firstly, according to agency theory, I hypothesise that dual-holders should mitigate shareholder-creditor conflict affecting the company they hold. I predict that the presence of dual-holders positively affects company value and performance facilitating distress resolution. This effect should be increasing in the number of dual-holders and their stakes in the company.
Secondly, I will elaborate on the presence of dual-holders with large blocks that can reduce agency conflict (Shleifer and Vishny, 1986) and cause such conflict with minority investors, including minority dual-holders, and negatively affect firm value (Thomsen et al., 2006). I hypothesise that block-dual-holders should mitigate shareholder-conflict around dis- tress but can cause it far from distress. I predict that the presence of block-dual-holders facilitate distress resolution and positively affect company value, but exaggerate conflict with minority investors far from distress pushing the company value down. Both effects should be increasing in the equity stakes of block-dual-holders.
Thirdly, I am going to estimate the impact of activist dual-holders on the company they hold. Activists usually face a free-rider problem requiring sufficiently high stakes to compensate the cost of activism (Grossman and Hart, 1980), but also activists are successful in initiating and influencing changes in the firm’s governance structure (Clifford and Lindsey, 2016). I assume that activist dual-holders would have a higher impact on the company than non-activist ones. The presence of activist dual-holders should have a positive impact on the company and positively facilitate distress resolution. These effects are expected to be growing in the number of activists and their equity/bond stakes. In the line with activist dual-holders, I tend to estimate the impact of passively managed dual-holders on the company they hold. Large concentrated passive stakes can overcome the free-rider problem by decreasing coordination costs of activism (Brav et al., 2008; Bradley et al., 2010), and amplify mitigation of shareholder-creditor conflict via more equal voting rights, more independent directors and long-term improvements of firm performance (Appel et al., 2016). The following hypothesis considers that passively managed dual-holders significantly impact the company than actively managed ones. I predict that the presence of passive dual-holders has a positive impact on a company far from distress, but this effect should disappear around distress due to the firm will be excluded from the index and passive dual-holders divest their stakes.
Fourthly, the crucial point in agency conflicts relates to the level of connectedness and business ties between investors and firm management. Voting of mutual funds is significantly influenced by their business ties with portfolio firms (Cvijanovic et al., 2016). I hypothesise that connected dual-holders are going to be less efficient in shareholder-creditor conflict resolution. The presence of connected dual-holders would cause a negative impact on the company, increasing with the number of connected dual-holders and their stakes. The last but not least point relates to dual-holders influence on company financing. Shareholder-creditor conflict is notable in debt overhang and constrained access to financing, however, dual-holders pro- mote improvements of credit conditions and easier access to debt financing (Jiang et al., 2010). I assume that dual-holders should alleviate concerns raised by obtaining more external financing.
The presence of dual-holders would increase company value for financially constrained firms and ones with debt overhang.
The preliminary results predominantly support the highlighted hypothesis and my predictions. Moreover, I attempt to address the issue of endogeneity by using: (1) mergers between mutual funds as an exogenous shock for ’dual-holder’ status; (2) instrumental variables for identification of block-dual-holders and activist dual-holders.
My research helps to understand whether dual-holders can alleviate shareholder-creditor conflict from many perspectives, e.g. size of investor stakes, monitoring activity, trading strategy and connectedness. It contributes a growing strand of literature regarding dual-holdings but also complements literature regarding agency theory and corporate governance.
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 approach. Our empirical experiments including portfolios based on five informative covariates (R-square, Return Gap, Active Weight, FundSize, Cash Flow) demonstrate truly positive performance, surpassing the portfolios of the latter and those based on sorting the covariates without controlling of luck. We conclude that the covariates carry valuable information in mutual fund selection improving investors’ performance.
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 Misreaction, Biased Beliefs, and the Mispricing Cycle
We construct a new measure that captures market misreaction to earnings information (“EMR”). High EMR scores predict return reversal consistent with an overreaction to firm information. EMR also positively predicts analyst forecast errors and firm mispricing (overvaluation) scores, and analyst forecast errors are slower to converge when EMR provides confirming information. In turn, EMR 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 misreaction feeds back into the belief formation of analysts, which partially explains the slow correction of firm
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": Common Lending and Loan Contract Structure
Consistent with common lenders internalizing industry spillovers arising from product market competition, I show they extend cheaper loans in exchange for stricter covenants, reducing borrower's competition appetite and externalities on other loans in bank's portfolio. Common lenders provide stricter capital-based covenants which align shareholder-debtholder incentives and deter excessive risk-taking "ex-ante", are more (less) likely to impose capex (payout) restrictions and restrain more at mature industries. Exploiting bank mergers, I verify findings are robust to endogeneity concerns or alternative explanations (e.g., specialization). Altogether, I present an explicit channel through which common lenders curb risky investment growth and thus maximize debt value at industry level.
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 Counterparty Default Insurance in the OTC Derivatives Market
This paper analyzes the effect of mandatory counterparty default insurance of over-the-counter (OTC) derivatives on aggregate financial risk exposure, focusing on the trade-off between decreased credit risk and increased market risk. I carefully model the competitive mechanisms in both the OTC derivatives market and their insurance market. I show how a for-profit central counterparty benefits from influencing the equilibrium outcome in the derivatives market, but fails to internalize the risk-mitigating objective. Mandatory counterparty default insurance therefore results in lower downstream competition, strictly higher prices and reduced hedging activities. Yet, this must be weighted against relatively safer sellers due to fewer uninsured derivatives sales and higher collateralization.
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 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 inevitable consequence of banks' maturity mismatch.
We 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 and higher loan proportions in their balance sheets.
However, we also confirm a remaining exposure to changing market rates.
Our results shed light on an implicit hedging mechanism within the traditional business model of banks, its (in)completeness, 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.
National Culture and Corporate Cash Holdings: Evidence from China
This article identifies national culture as an important factor affecting corporate cash holdings by using China and its national culture, Confucianism, as a setting. We find that firms located in regions with higher levels of Confucian culture hold higher levels of cash and this high cash holdings status is also more persistent. Next, we employ an instrumental variable to establish causal identification of the culture effect and the IV estimates show a compelling economic magnitude. The effect represents a 12% increase in corporate cash holdings from the sample average if the regional Confucian culture density is one standard deviation higher. 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 find that higher Confucian culture firms make better investment decisions, have higher acquisition announcement returns, pay out more dividends and achieve higher profitability and lower profit volatility. These all are additional evidence arguing against agency motives, but supporting an efficient outcome argument. Finally, we also show that the CEO/board chairman's Confucian background and the firm's headquarter cultural environment together exert an influence.
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
This paper examines the importance of protected areas for weather disaster mitigation and nature's impact on local economies. Specifically, I investigate how shocks to local natural capital affect municipal bond markets. Using 313 natural capital loss events and extreme weather events, I show that counties with protected areas face lower natural disaster damages and lower cost of debt. The additional damages from natural disasters are estimated at between $9 and $23 million. Moreover, the additional cost of debt related to the loss of natural capital can be as high as $690,000 for an average bond. Lastly, the results show one of the social costs related to the loss of natural capital: population migration. Overall, this paper provides evidence supporting nature conservation and highlights the important role of nature as a shield from natural disasters.
New Cryptocurrency Indices
We have developed and made available a new Cryptocurrency Uncertainty Index (UCRY) based on news coverage. Our UCRY Index captures two types of uncertainty: that of the price of cryptocurrency (UCRY Price) and uncertainty of cryptocurrency policy (UCRY Policy). We show that the constructed index exhibits distinct movements around major events in the cryptocurrency space. We suggest that this index captures uncertainty beyond Bitcoin, and can be used for academic, policy, and practice-driven research.
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.
Opioids Epidemic and Mortgage Default
I examine the impact of the opioids epidemic on households' defaults using census tract-level conventional mortgage performance data in the US. I find that higher rates of opioids overdose mortality rates are associated with higher delinquency rates among prime borrowers. Using instrumental variable and difference-in-difference estimation strategies, I establish a causal link between the prescription opioids crisis and mortgage default rates. The role of labor income loss is trivial to explain this causal link. I document that home equity is the dominant economic channel through which the opioids crisis affects mortgage defaults.
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 exploits China’s recent anti-corruption campaign to study the effects of political connection on corporate litigation. Specifically, by exploiting the enforcement of an important component of the campaign, the Communist Party of China’s Rule 18 which forces politically connected independent directors to resign from public firms, we investigate how acquired protection from political connection may distort corporate litigation. We show that the weakening of political connection results in higher likelihood of and larger amount involved in corporate litigation against connected firms. Such effects are more pronounced for non-state-owned enterprises, financially distressed firms, and firms in regions with worse legal institutions. We also find that cases with severe information asymmetry are a crucial driving force of the effects. Finally, additional tests suggest that the weakened connection is associated with higher win rates of plaintiffs and more negative market reactions to corporate litigation.
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.
Real Effect of Bank's Block-holding on Firm's Market Power
This paper studies the link between bank’s equity holding of non-financial firms and markups of firms. We exploit bank mergers as exogenous shocks to banks' block-holdings of top rivals and show that rival-creation bank mergers lead to lower markup (gross profit margin), and higher probability that firms switch bank. These effect are stronger when an industry is highly competitive, when a firm is R&D intensive, and when the bank has more private information of the firm, which is consistent with proprietary information leakage hypothesis. Besides, we find that bank-held rivals enjoy higher market power in the product market.
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 seven well-known predictors for fund performance from the investors’ perspective. Using a recursive out-of-sample procedure, I demonstrate a discrepancy between ex post and ex ante predictability. More importantly, the investors’ flow response to predictor-implied performance exhibits strong variations across predictors, suggesting that investors may use information embedded in predictors for fund selection even though they cannot benefit from them in real time. These findings support the dumb money effect of mutual fund investor flows.
Ruling with Ideology: Politician Belief and 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. I find that other things equal, pro-communism party secretaries (governors) privatize 13% (12%) fewer firms than less communist secretaries (governors). I also find that party secretaries influence privatization mainly through subordinate government-controlled firms, while governors also affect firms they directly control. 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?
We present a model-free approach to recover the serial dependence in the stock market from the publicly-traded derivatives. By using S&P 500 index options, VIX futures, and VIX options, we quantify the market autocorrelation and time-series regression coefficients in real time. Our empirical results suggest a negative reaction (resp. AR(1) beta) ranging from –20% to –40% between two consecutive market returns, but the reversal level is relatively small compared to the past, yielding an upward market trend in the long-run (resp. AR(1) intercept term). We further illustrate that the derivative-inferred autocorrelation is comparable to the realized month-to-month sample correlation using historical data, and we show that learning past information in stock market with fading memory may reconcile real time forward-looking derivatives.
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.
Shrinking Boundary of the Invisible Hand
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 positive ESG premium among stocks with a low degree of ESG-motivated investor ownership. ESG-motivated investors buy ESG stocks giving them high ex-ante but low ex-post abnormal returns. We show that a theory of sustainable investing with heterogeneous skill and sustainability sentiment can explain this finding. In support of this explanation, we find in the cross-section that a low degree of ESG-motivated ownership leads to future ESG score increases. The premium is stronger during periods of high climate sentiment and risk aversion as in the crisis.
Social Investing and Hype in the Stock Market
Using real-time retail group chatter from social media, I find that social investing provides investment value. I show that Discord demonstrates smart chatter while Reddit is better as a leading indicator for meme stocks. I create a hype measure that positively predicts trading volume, stock volatility, and future returns, tilting towards small, aggressive, growth, low- profitability, and loser stocks. I find that higher returns are driven by continuity of hype than by the days stocks become hyped, and that insiders are less likely to sell after they become hyped. Moreover, the returns continuously drift upwards, suggesting that 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 well-documented distress anomaly results, in part, from a reduction in firm risk following the execution of a credit forbearance agreement with firm creditors. 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 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.
The Effect of Bank Competition on Deposit Price Dispersion
We 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, we show that the local market’s bank concentration has a negative effect on the price dispersion of the deposit products. We further points out that this negative effect holds for the different types of deposit products (e.g. Certificate of Deposit, Money Market deposit). However, this negative effect of the the local market’s bank concentration on the price dispersion attenuates with the increase of the maturity period of the deposit products. The negative effect of the bank concentration on the deposit price dispersion hints about the existence of the price inelasticity in the deposit market and implies that depositors might prefer other factors (e.g. distance, convenience) over pricing in their decision making about choosing a bank.
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 and the valuation of non-deal peer firms. Specifically, the information acquisition about merging firms enhances the relation between combined announcement-period abnormal returns and post-merger operating performance of the combined firm. The effects are stronger in deals that experience high institutional downloads and high trading volume. Furthermore, information acquisition in peer firms around M&A announcements strengthens the link between their short-term abnormal stock returns and long-term operating performance. Non-deal firms with greater download activity experience an increase in price informativeness and subsequent takeover probability. Overall, this paper provides supportive evidence that information acquisition improves the market’s assessment of the merger synergies and valuation on merger-related firms.
The Evolution of Market Efficiency Over the Last Century
Combining a novel hand-collected sample of earnings announcements from the Wall Street Journal archives with more recent data from Compustat, I document a striking inverted U-Shape in the nature of market efficiency over the period, 1934 - 2018. In terms of investors’ response to earnings announcements, markets are more efficient both early and late during this period, while they become less efficient during the middle portion. I further show that higher efficiency in early and later periods are driven by two different economic dynamics. While speed in information processing lead to more efficiency in the later periods, the surprisingly high degree of market efficiency in the 1930s and 1940s likely reflects the relative importance of earnings announcements as a central source of information for investors, when there was less overall information to process and fewer other information venues to focus on. Overall, my results highlight that the evolution of market efficiency has not followed a linear path but rather divergent economic forces caused the inverted U-Shape pattern.
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 relies on the industry expertise a bank gains through lending to firms in a specific industry. The industry expertise strengthens the bank’s understanding of economic conditions in counties where the industry is a major sec- tor. Thus, for households in those counties, banks can better evaluate their short-term and long-term financial health, and hence their mortgage affordability. Information gained from the channel improves banks’ screening and monitoring efficiencies, leading to increased allo- cation of mortgage credits towards counties with same industry specialization. The effects are stronger when the information asymmetry between banks and mortgage borrowers is high or when local risk is high. Further tests show that mortgages originated through the channel contain more soft information and have better performance.
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 then analyse the effects of Trump tweets on the intraday trading activity of foreign exchange markets, such as trading volume, volatility and FX spot returns. We find that Trump tweets reduce speculative trading, with a corresponding decline in trading volume and volatility, and induce a bias reflecting Trump’s (optimistic) views on the U.S. economy. We rationalise 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.
The Real Effect of Competition Laws: International Evidence
In this paper, we examine the effect of competition law which regulates the operation of markets by restricting anti-competition actions on firms’ financing and investment outcomes. Using a comprehensive sample of about 207, 080 firm-years observations from 78 countries, we find that competition legislation increases firms access to external financing and leads to more investment. This finding is robust to an array of robustness tests including employing different sample, using different fixed effects, and using alternative clustering methods. In cross-sectional tests, we find that the positive association between competition law and corporate financing and investment is stronger for firms in economies with weaker investor protection laws and for less financially constraint firms. We also find that the authority sub-component of the competition law index has a more pronounced effect on firms’ financing and investing decisions than do the substance sub-component. By using a sample that covers a wide of range of industries and countries, our study help broaden our current understanding of the consequences of product market competition, in particular how variation in institutional settings matter for the success of competition law 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 usually assume that price and quantity are continuous variables, while most market designs, in reality, impose discrete tick and lot sizes. We study a firm’s trade-off between these two discretenesses in U.S. stock exchanges, which mandate a one-cent minimum tick size and a 100-share minimum lot size. A uniform tick size favors high prices because the bid–ask spread cannot be lower than one cent. A uniform lot size favors low prices because low prices reduce adverse selection costs for market makers when they have to display at least 100 shares. We predict that a firm achieves its optimal price when its bid–ask spread is two ticks wide, when the marginal contribution from discrete prices equals that from discrete lots. Empirically, we find that stock splits improve liquidity when they move the bid–ask spread towards two ticks; otherwise, they reduce liquidity. Liquidity improvements contribute 95 bps to the average total return on a split announcement of 272 bps. Optimal pricing can increase the median U.S. stock value by 69 bps and total U.S. market capitalization by $54.9 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 the credit markets. The empirical setting exploits the outburst of the Wells Fargo scandal as a negative shock to the 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 more trust erosion has a larger increase in FinTech share relative to a county experiencing less trust erosion. Treatment effect heterogeneity estimations from both regression and generic machine learning inference suggests that trust is less critical in FinTech adoption for African American borrowers.
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 in 17 Eurozone countries from the ECB Household Finance and Consumption Survey (HFCS) and look at the household credit in the ”pre-APP (ECB’s Asset Purchase Programmes policy)” period and the ”post-APP” period. I use weighted least squares regression and recentered influence function regression joint with the Oaxaca-Blinder decomposition method to identify two potential channels of unconventional monetary policy on household credit inequality. (1) The credit bias channel increases the credit inequality between the top and the bottom quintiles of the income and wealth distribution, because the APP policy makes households with assets wealthier and more welcomed by the banks. (2) The credit constraint channel reduces credit inequality and benefits the low and middle wealth quintiles, as previously credit-constrained households increase relatively more their refinancing and new mortgages for house purchases after the APP. Moreover, there is country heterogeneity in policy transmission. The credit constraint channel works stronger in peripheral countries than in core countries.
US Wealth Shares, the Dollar and International Risk Sharing
I study the joint dynamics between the US wealth share, the dollar and the global economy. I uncover two novel stylised facts about these joint dynamics. Firstly, the US wealth share is countercyclical: it tends to rise during global bad times. Secondly these wealth share fluctuations are not driven by the dollar: they are generated by local equity valuation forces associated with relative US stock market outperformance during global bad times. These facts present a natural challenge to modern international macro-finance models that place the US and the dollar at the center of international risk sharing and global risk pricing. I rationalise these joint dynamics using a simple two country recursive model of international risk sharing.
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.