AFA Ph.D. Student Poster Session
Poster Session
Sunday, Jan. 3, 2021 7:00 AM - 6:00 PM (EST)
Monday, Jan. 4, 2021 7:00 AM - 6:00 PM (EST)
Tuesday, Jan. 5, 2021 7:00 AM - 3:00 PM (EST)
Access to Banks and Household Resilience
Abstract
This paper investigates how increased bank presence in rural villages affects households' ability to cope with adverse shocks. I use a nationwide policy of the Reserve Bank of India from 2005 that induces exogenous variation in bank presence and combine it with a nationally representative household-level survey. The policy incentivizes banks to set up new branches in 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 below and just above the national average. In absence of complete insurance, households use bank savings and bank loans to cope with adverse events such as droughts or health emergencies. For aggregate and idiosyncratic shocks respectively, I investigate how improved resilience through banking services impacts overall welfare and discuss mechanisms. Additionally, I examine how access to banking services influences take-up of informal and semi-formal financial services as well as over-indebtedness of households.Are Green Investors Green-Inducing? A Demand System Approach
Abstract
Despite the growing interest in green investing among academics and industry professionals alike, there is little consensus on whether it successfully incentivizes firms to adopt eco-friendly business practices. Using the equity holdings of institutional investors and the “demand system approach” to asset pricing, we provide evidence that institutional demand for greener stocks encourages firms to improve their environmental performances. Specifically, we devise and estimate a firm-level quantity, institutional pressure for greenness, that measures the price pressure a firm receives from its institutional owners. We find that this quantity has a positive and significant relationship with future improvement in a firm’s environmental performance. Together with results from placebo tests, we conclude that green investors, those with high portfolio-level environment scores, are not necessarily green-inducing investors, those who encourage better en- vironmental performance. Instead, green-inducing investors are institutions who contribute to higher institutional pressure, i.e. investors who are price-inelastic and display a positive portfolio tilt towards greener assets.Banks with Diagnostic Expectations
Abstract
We analyse the loan loss provisioning behaviour of the banks in United States and find evidence for departure from rational expectations. Banks, while estimating future losses on their respective loan portfolios, tend to overreact to the current observed losses and make systematic errors in their forecasts. We show that these forecast errors and the resultant expectation bias at the level of bank can explain credit growth as well as cross section of equity returns in the immediate one-three year period; for instance, banks with an optimism bias, witness a decline of 3.7% in the one-year ahead predicted return. We model the role of such biased expectations in a DSGE setup and show that the presence of bias in the banking sector: (i) adds to procyclicality of credit supply (ii) reduces the effectiveness of the recently introduced dynamic provisioning rules and incentivises banks to take excess risk in response to it. The trade-off between the benefit of reduced procyclicality and cost of excess risk becomes more intense with the extent of bias.Benefiting from Bankrupt Firms: Evidence from Local Labor Movement
Abstract
A bankrupt firm can release many of its employees, who may subsequently be recruited by firms operating in the same geographic area. We examine the long-run effect of bankruptcy announcements on stock performance of companies that are located in the same city as the bankrupt firm. Using event study and time-series analyses, we find positive abnormal returns in the three years after the bankruptcy filing. Further tests attribute this effect to labor movement. For example, companies in the same city have higher abnormal returns when they add more employees after the bankruptcy, and when the bankrupt firm has more employeesBetter Be Careful: The Replenishment of ABS Backed by SME Loans
Abstract
We investigate the replenishment of 102 asset-backed securities (ABS) backed by more than 1.7 million small- and medium-sized enterprise loans. Based on our extensive data set from 2012 to 2017 obtained from the first and only loan-level central repository for ABS in Europe, we reveal that loans added to securitized loan portfolios after the transactions' closing perform worse than loans that are part of the initial portfolio. We additionally provide evidence that originators induce these performance differences since they exploit their information advantage by deliberately adding low-quality loans to securitized loan portfolios. This adverse behavior is mitigated by originators' reputation efforts, by increasing transparency in the ABS market, as for example per the European Central Bank's loan-level initiative, and most effectively by their interaction.Boosting Portfolio Choice in the Big Data Era
Abstract
Markowitz’s mean-variance portfolio optimization is either inefficient or impossible when the number of assets becomes relatively large. To overcome this difficulty, we propose several component-wise boosting learning methods that, in a linear regression specification, can iteratively select the assets (variables) with the largest contribution to the fit from a huge number of assets. Based on dataset consisting of 897 assets with 624 observations obtained from Ken French data library, we assess the performance of tangency portfolios estimated using our methods. We find that our methods substantially outperform the 1/N portfolio in terms of various popular metrics. For example, our component-wise LogitBoost can reach an out-of-sample Sharpe ratio of 1.03, while the 1/N portfolio only achieves a Sharpe ratio of 0.27.Intermediated Credit and Local Resilience
Abstract
-We test whether bank capitalization affects local resilience during crises.-We exploit a shock to the real economy from the COVID-19 pandemic.
-We find that counties with poorly capitalized local banking sectors exhibit more business closures, more unemployment, and more declines in income and hours worked during the pandemic.
CDS Central Counterparty Clearing Default Measures: Road to Recovery or Invitation to Predation?
Abstract
Post the 2008 financial crisis, regulation demanded the standardisation of credit default swaps (CDS) and their mandatory clearing through Central Clearing Counterparties (CCP). Hence, CCPs have become systemically important institutions, whose potential failure poses a serious threat to global financial stability. This work examines the potential failure of a CCP when required to un- wind the positions of a large dealer bank, following default. The model captures the price impact of liquidation and predatory selling through the exchange of variation margin between dealer banks. It provides a novel measure of covariance between banks’ CDS holdings. Key results show that liquidation lowers CCP profits, and that predation decreases the profits of all members, pushing banks to default. Implementing a hybrid CCP (vs. current) structure provides a natural disciplinary mechanism for predation. A dynamic simulation based on OTC market data provides parameter sensitivities for the magnitude of gains/losses faced by the CCP and predatory banks. These are minimised with a hybrid fund structure. Finally, under various market liquidity scenarios, regulatory implications are obtained for the timing of liquidity injection by a Lender of Last Resort (LoL).Collateral, Haircuts and Rates: A Theory of Repo
Abstract
We study the effect of the quality of collateral on repo rates and haircuts. We build a model of repo based on differences in beliefs, where Value-at-Risk and Expected Shortfall arise endogenously as sufficient statistics of the quality of collateral, i.e. its return distribution. Although a higher Expected Shortfall increases both haircuts and repo rates, a higher Value-at-Risk leads to a larger haircut and a lower repo rate. We also show that although riskier borrowers face higher haircuts, they do not necessarily pay higher rates. More profitable borrowers choose to borrow more against the same collateral at a cost of paying higher rates.Risky Financial Collateral, Firm Heterogeneity, and the Impact of Eligibility
Abstract
How does the eligibility of risky corporate bonds as collateral in Central Bank Operations affect financial intermediation, corporate financial policy and aggregate economic activity? While a large strand of literature suggests that relaxing collateral constraints on financial markets is beneficial (the financial market channel), a thorough assessment of collateral policies must also account for endogenous responses of the non-financial sector (the firm channel). The latter channel arises because banks are willing to pay price markups on corporate bonds, if they can be used as collateral, and firms heterogeneously adjust their borrowing and dividend policies. To jointly analyze both channels, we construct a model with a collateralized interbank market and heterogeneous firms that may default on their debt. In this setup, price markups from collateral service distort the trade-off between tax advantages of leverage and bankruptcy costs. In an analytically tractable version of the model, we demonstrate that high-risk firms are more likely to reduce leverage while low-risk firms venture into riskier borrowing regions. To the best of our knowledge, the redistributive consequences of this margin of Monetary Policy have not been described so far. Since the aggregate effects are difficult to account for analytically, we calibrate the model to the Euro Area, generate a simulated panel of firms, and find effects of eligibility consistent with several empirical studies. With our calibration, we quantify the effects of relaxing eligibility standards in response to Great Financial Crisis of 2008 at 5bp, relative to a counterfactual scenario without policy relaxation. Key to this relatively small effect is the simultaneous increase of corporate default risk and tightening of interbank markets.Common Ownership along the Supply Chain and Supplier Innovations
Abstract
Common owners are the (institutional) investors that hold equities of multiple firms. This paper examines the impact of common ownership of suppliers and customers on suppliers' innovation activities. I find suppliers' investment in innovation, quantity and quality of innovation output increase when common owners control higher fractions of the suppliers and their customers. The impact of vertical common ownership on innovation input and quality of innovation output is stronger and more robust than that of horizontal common ownership. I provide plausible evidence for causality using a difference-in-differences approach based on a quasi-natural experiment in the form of financial institution mergers and acquisitions. Moreover, I test the potential channels through which vertical common ownership could influence supplier innovation. My evidence suggests that common ownership increases investment in innovation by mitigating hold-up issues between suppliers and customers, and enhances innovation output performance by improving technological spillovers between suppliers and customers. Overall, my evidence suggests that common institutional ownership enhances suppliers' innovation performance by improving relationships between suppliers and their customers.Corporate Innovation in the Cyber Age
Abstract
We construct and validate a text-based metric capturing firms' ex-ante exposure to cybersecurity risk, and we document that the rise of cyber threats is redesigning corporate innovation strategies. As firms' exposure to cybersecurity risk increases, managers' reliance on trade-secrets declines, as they seek to protect their firm's intellectual capital under patent and intellectual property laws. Besides increasing their patenting activity, we document that firms exposed to cyber threats file for simpler patents to accelerate their innovation cycle. Finally, we show that this strategic adjustment is not costless, as it causes firms' returns to R&D investments to decline significantly.Corporate Innovation Linkages and Firm Boundaries
Abstract
Innovation matters for firm boundaries. Companies are more likely to integrate with peers with connected innovation. In this paper, I study how follow-on innovation determines the degree of integration between firms. I construct a measure of relative innovation proximity between firms, based on patent citations. I find companies are more likely to acquire peers with closer follow-on innovation, rather than build strategic alliances with them or license/buy their patents. Furthermore, the measure of relative innovation proximity between firms reflects firms' bargaining power and not the size of the synergies. In M\&A transactions, a bidder with closer follow-on innovation pays a greater premium and exhibits lower announcement returns. On the other hand, in strategic alliance, a firm with closer follow-on innovation experiences greater announcement returns. These results are consistent with a hold-up model in which companies bargain over the type and terms of the contract.Corporate Social Responsibility and Market Efficiency: Evidence from ESG and Misvaluation Measures
Abstract
We study the impact of corporate sustainability on market efficiency in the US. Our results indicate that a firm’s Environmental, Social and Governance (ESG) profile significantly affects valuation: an enhancement of a firm’s corporate sustainability leads to a higher ratio of actual to true firm value. Analyzing the relation between ESG and misvaluation separately, we find that ESG expands existing overvaluation whereas it reduces undervalued firms' deviation from the true value. We argue that the extension of overvaluation is driven by the trend to invest sustainably. The reduction in undervaluation, however, is attributable to increasing information availability to capital markets. Further analyses reveal a moderating role of market sentiment towards sustainability in the ESG-misvaluation relationship.Cost of Conscious Capitalism
Abstract
Calls for conscious capitalism have led to numerous innovations in the governance of firms. In this paper, we try to assess whether there is a cost to investors for one such governance innovation—the introduction of constituency statutes at the state level within the United States. Constituency statutes expressly permit board members and management to be more mindful of the interests of all stakeholders—not just equity shareholders—when making decisions. We argue that as the interests of competing stakeholders make demands of corporate decision-makers, those decision-makers will try to reduce the possibility of being held accountable—by reducing transparency altogether. Using a sample of U.S. publicly traded firms (1981-2010), we observe a significant decline in transparency by firms incorporated in states with such statutes. However, while we find that firms going through losses use conscious capitalism as an umbrella to remain opaque, the firms that need financial markets for capital remain transparent even after the adoption of statutes. Overall, our results support the proposition that the adoption of constituency statutes negatively impacts corporate transparency and accountability towards shareholders, weakening governance.Dark Trading and Financial Markets Stability
Abstract
We examine how the implementation of a new dark order type – Midpoint Extended Life Order (M-ELO) on NASDAQ – impacts financial markets stability in terms of occurrences of mini-flash crashes in individual securities. We use high-frequency order book data around the implementation date and apply panel regression analysis to estimate the effect of the dark order trading activity on market stability and liquidity provision. We find that the introduction of the M-ELO increases market stability by reducing the average weekly number of mini-flash crashes and improves the liquidity provision in terms of spreads, and available depth.Data Security and Merger Waves
Abstract
We examine whether data security affects the merger waves in the era of big data. Exploiting the staggered adoption of laws that require mandatory disclosure of data breaches across states, we find that the intensity and likelihood of M&As increase (decrease) in states of targets after the law was adopted, when acquirers are from a state with (without) the laws in place. The increase of M&As was contributed by the mitigation of data lemon problems as a result of enhanced cyber security. The decrease of M&As was due to higher costs associated with potential data breaches. The effects are stronger among industries that are more competitive and data intensive. Overall, our findings highlight the importance of cybersecurity in the era of big data and digital economy.Decomposing Factor Momentum
Abstract
Factor momentum returns do not stem from momentum in factor returns. To study the source of returns, this paper decomposes the stock factor momentum portfolio into a factor timing portfolio and a static portfolio, where the former dynamically collects the return due to serial correlations of factor returns and the latter passively collects factor premiums. Sequential sorts of 210 factors by factor characteristics show that the static portfolio robustly accounts for a dominant fraction of the factor momentum return and outperforms in risk-adjusted returns, whereas factor return predictability is empirically too weak to produce timing benefits. The static portfolio survives the post-publication decline of factor performance but the factor momentum portfolio does not.Direct Listing or IPO? The Anatomy of the Going-Public Market
Abstract
This paper analyzes the impact of direct listing (DL) innovation on firms’ going public decisions and the welfare consequences. Under adverse selection, DL and IPO markets cater to different types of firms. DL market is more vulnerable to breakdown. Imposing certification intermediaries is essential in maintaining a well-functioning DL market, which leads to more firm entry into the public market and improved social welfare. DL firms and intermediaries enjoy welfare gains, while public investors may face higher risks. The model rationalizes firm heterogeneity in the U.K. DL and IPO markets, low firm participation in the U.S. DL market, and investment banks’ support for DL. The paper implies that better-developed private capital and stock trading markets motivate DL innovation. The paper also highlights the benefits of going-public other than capital-raising, the severe informational frictions in the going-public market, and the importance of regulation in protecting public investors and preventing market failure.Do Directorship Returns Affect CEO Turnover?
Abstract
Boards staffed by directors who have experienced high stock returns at other firms where they hold directorships are more likely to fire their CEOs. A one-standard-deviation increase in such directorship returns increases the probability of forced turnover by 11-19%. Directorship returns have a larger impact when they are more recent and when CEO ability is harder or costlier to evaluate, consistent with boards using an availability heuristic to form beliefs about the ability gap between incumbent and replacement CEOs. Conditional on forced turnover, directorship returns are inversely correlated with subsequent improvements in operating performance, suggesting that such beliefs are suboptimal.Distorted Beliefs and Exchange Rates
Abstract
In a risk-neural world, the equilibrium exchange rate is determined by the Uncovered Interest Rate Parity condition which equalizes the exchange rate to the current interest rate differential between the domestic and foreign countries. This implies that a regression of realized excess returns of investing on foreign short-term bonds on current interest rate differential should produce a coefficient of zero. The economic reason is that if the domestic (foreign) short-term interest rate increases above the foreign (domestic) interest rate, we should expect the domestic (foreign) currency to depreciate. To put it another way, the market's expectation of currency returns should be equalized, so that excess returns are not anticipated ex-ante and should not be predictable. Nevertheless, Fama (1984) and many other subsequent empirical tests rejected a coefficient of zero and the results are quite robust running on different time-series and currency-pairs. These lead to several well-documented major exchange rate puzzles. For example, a positive regression coefficient in short term implies that currency with higher interest rate has higher expected returns (forward-discount puzzles) and a negative coefficient in long run however suggests that currency with higher interest rate has lower expected returns after several periods (predictability reversal puzzle). Combing these two puzzles results in the delayed overshooting puzzle. Although the prior literature built on standard monetary models have successfully provided a partial explanation for some of the anomalies, they can hardly be applied to jointly explain all the major puzzles within one framework. In addition, the past papers are insufficient to provide quantitative estimates. Along with the recent popularity of directly using survey forecast data, this paper develops a new behavioral monetary model to study the behavior of equilibrium exchange rate. Furthermore, we make use of professional forecast data of short-term interest rate to proxy market expectation and calibrate the model to uncover a range of model parameters in order to conduce numerical exercise. In this paper, we build a heterogeneous-agent model with private information. The interest rate differential between domestic and foreign country is determined by an exogenous process. There is a latent economic variable which has an AR structure and the realized interest rate differential is a linear function of the underlying fundamental variable and a transitory shock component. The fundamental variable which governs the interest rate process is not observable to any agents, but we allow each agent to observe a private signal of it. Upon seeing the private signal and the current interest rate differential realization, each agent updates her belief of the latent variable in a Bayesian fashion by a signal extraction task. The beliefs of all agents are then aggregated to market expectation which determines the equilibrium results. In the mode, we do not assume agents are fully rational. Deviating from rational benchmarks, we instead assume that agents do not use the true model parameters when updating beliefs. This is motivated by the empirical evidence of the dynamic responses of forecasts to the interest rate shocks. Following Kucinskas and Peters (2019), we estimate the impulse response function of forecast errors. The results are consistent to the notion developed in Angeletos, Huo and Sastry (2020): under-shooting (under-reaction) early on but overshooting (over-reaction) later on in response to aggregate shocks. These patterns are also helpful in explaining the empirical puzzles identified in the literature. For instance, the forward-discount puzzle and the predictability reversal puzzle are consistent to investor under-shooting and overshooting later on respectively. As a result, to reconcile the empirical regularity, we assume that agents in our model perceive the AR process differently from the true process. In addition, the agents also misunderstand the precision of their private signals. When the agents think the underlying fundamental is more (less) auto-corrected, the agents over-extrapolate (under-extrapolate) the time series of realized interest rate differential. And if an agent believes her signal is more (less) precise than the reality, she is defined as overconfident (under-confident). Our results show that over-extrapolation combined with slow learning (the perceived precision is not too large) is able to generate the estimated impulse response function which switches sign after some data periods. To understand the intuitions, we can think of two economies where the interest rate differential is zero before a given period then a negative shock occurs. We can simply assume there is no additional shocks aftermath. Without loss of generality, we can take US dollars (domestic) and Canadian dollars (foreign) as a concrete example. In the initial period, the interest rate in Canada experiences a positive shock when the interest rates in the two economies are the same before this period. Due to slow information learning, the agents do not fully appreciate the negative shock to the interest rate differential (US interest rate - Canada interest rate). This causes the initial under-shooting in forecasting. As a result, she over-invests in the foreign bond market which causes the home currency depreciates less than it should be. This further implies positive excess returns. After a sufficient time period of learning, the agents are aware of the interest rate shock. But the agents over-estimate the negativity of interest rate differential due to over-extrapolation. Hence they over-invest in the US market, resulting in over-appreciation of the US dollars and reversal predictability (negative excess returns). Finally, we take the model into data. In the model, the expectation of the unobserved fundamental in each period can be easily derived, therefore also the market expectation of interest rate differential. This is essentially the theoretical equation we estimate using forecast data. The data we rely on are the U.S. interest rate from 1985 to 2018. We first estimate the true process of the interest rate differential by which the AR coefficient and the precision of the shocks can be recovered. Our distributional assumptions on the random shocks result in a standard state-space process. Hence, we conduct a simple MLE exercise to estimate the behavioral parameters from the investor side. Our empirical results show the calibrated model is successful in matching the UIP regression coefficients and the IRF of the forecast errors.Do Banks’ Partisan Affiliations Shape their Lending Decisions?
Abstract
This paper provides novel evidence on the role of banks’ partisan affiliations (independent of political pressure or rent seeking) in their lending decisions. Using county-level data on mortgage applications over the period 1994 to 2017, we find that banks’ mortgage lending is consistently larger in counties that share their political beliefs. This finding is robust to alternate definitions of political affiliations of banks and counties and controlling for observable loan characteristics and bank-year and county-year fixed effects. We find no evidence that rent extraction or political influence can explain this difference in mortgage lending. Instead, ideological differences across banks based on their partisan affiliations seem to drive the results. We show that lending decisions based on partisan allegiances have a negative effect on banks’ health as reflected in their higher non-performing loan ratios and lower return on assets.Do Mutual Funds Manipulate Star Ratings? Evidence from Portfolio Pumping
Abstract
This paper reveals that mutual fund managers manipulate Morningstar ratings by inflating their month-end portfolio values when they are likely to finish the month near rating cutoffs. This star rating manipulation is more pronounced among funds with greater incentives and abilities to pump their portfolios and manipulate star ratings. Following heightened regulatory scrutiny, portfolio pumping to manipulate star ratings has largely migrated from quarter/year-ends to less prominent month-ends. Improving star ratings, portfolio pumping increases fund flows, especially in the month of a rating upgrade. Placebo tests exploiting the June 2002 change in Morningstar's rating methodology yield expected null effects.Do Sell-Side Analysts Say “Buy” While Whispering “Sell”?
Abstract
I examine how sell-side equity analysts strategically disclose different information to the public and buy-side mutual fund managers to whom they are connected. I specifically test the possibility that analysts’ public recommendations tell the public to “buy” but they are whispering “sell” to their connected fund managers. I measure the likelihood of such “say buy/whisper sell” behavior based upon the percentage of managers’ selling stocks that analysts recommend buying. Using mutual fund managers’ votes for sell-side analysts in a Chinese “star analyst” competition as a proxy for managers’ evaluations of analysts, I find that managers are more likely to vote for the analysts who exhibit more say buy/whisper sell behavior with these managers. This result suggests that managers receive more-precise information in private communications with an analyst than in the analyst’s public recommendations and reward the favor by voting for the analyst in the “star analyst” competition. This different information disclosure by analysts results in a form of information asymmetry, which incurs a significant cost on uninformed investors; among analysts’ positive recommendations, the stocks bought by the managers who vote for the analysts outperform the stocks sold by these managers around the recommendation dates.Does Corporate Diversification Retrench the Effects of Firm-Level Political Risk?
Abstract
This study investigates the effects of firm-level political risk on corporate investments and operating performance. We find that diversified firms are better able than focused firms in mitigating idiosyncratic political risk. Diversified firms accomplish this feat via efficient use of the internal capital market that allows segments to alleviate the adversity of political uncertainty. When exposed to political risk, diversified firms do not spend more on lobbying and political donations than the focused firms in the subsequent period, implying that diversified firms do not manage political risk politically. Our main findings are robust to a battery of endogeneity tests.Does Forecasting Price Efficiency (FPE) Affect Revelatory Price Efficiency (RPE)?
Abstract
This paper offers evidence that forecasting price inefficiencies signaled by corporate events affect stocks' revelatory price efficiency (RPE). RPE decreases after over-valuation signals, more in firms with worse investment opportunities, poor corporate governance, more entrenched managers, and higher short sale constraints. Whereas, RPE increases after under-valuation signals, more in firms with better investment opportunities and managers who listen to prices and during boom times. Results are stronger when Q and Price-to-Value and corporate events suggest mis-valuation in same direction. The results imply that market over-valuations are corrected slower and hence are stickier and more prevalent in the economy than under-valuations.Does Limited Liability Matter? Evidence from a Quasi-Natural Experiment
Abstract
We use the enactment of limited liability legislation across Canadian provinces to examine the effect of the change in liability status on firm outcomes for a group of Canadian public firms known as income trusts. We show that the switch from unlimited to limited liability increases the trusts’ net external financing, investments, profitability, payouts, and equity volatility. Our results are stronger for energy trusts, which are more capital-intensive. Furthermore, our event study results show significantly positive cumulative abnormal returns around the introduction of limited liability legislation. Overall, we present a novel approach to test the impact of limited liability on firms.Does Private Equity Ownership Make Firms Cleaner? The Role of Environmental Liability Risks
Abstract
Private equity (PE) ownership leads to a 70% reduction in the baseline rate of toxic pollution. The reduction is identified from the oil and gas industry using a nearest-neighbor research design estimated on novel satellite imaging and administrative datasets. I test several mechanisms that could explain this behavior. PE ownership's impact on pollution is negatively related to plausibly exogenous increases in regulatory risks, contrary to what either a non-pecuniary or technological upgrade channel would predict. Exploiting specific private equity deals from the energy industry, I find that PE control and incentive to sell the company are the main drivers behind the results. Additional tests support the view that reducing toxic pollution maximizes PE exit value by making the portfolio company attractive to more buyers.Does Social Connectedness Affect Stock Market Participation?
Abstract
Using IRS tax filing data, I show that social network and word-of-mouth communications play an important role in stock market participation decisions. Using a novel dataset from Facebook, I construct a measure of social network friends' participation for US counties and find that a one-percentage point increase in friends' participation increases the focal county participation by 14 to 25 basis points in the following year. For identification strategy, I employ the revelation of financial misconducts as an exogenous negative shock to local participation rate and show that the instrumented change in friend participation significantly and positively predicts the change in focal county participation rates. The increase in participation rates among the low-income households induced by friends' participation decreases the Gini coefficients in metropolitan counties in the following two years. The evidence suggests that social influences and peer effects contribute to the cross-sectional differences in the stock market participation rates across US counties and may lead to lower income inequality.Ask BERT: How Regulatory Disclosure of Transition and Physical Climate Risks affects the CDS Term Structure
Abstract
We use BERT, an AI-based algorithm for language understanding, to decipher regulatory climate-risk disclosures and measure their impact on the credit default swap (CDS) market. Risk disclosures can either increase or decrease credit spreads, depending on whether disclosure reveals new risks or sharpens the signal and decreases the uncertainty. Training BERT to differentiate between transition and physical climate risks, we find that disclosing transition risks increases CDS spreads, especially after the Paris Climate Agreement of 2015, while disclosing physical climate risks leads to a decrease in CDS spreads. These impacts are statistically and economically highly significant.Does the Internet Replace Brick-and-Mortar Bank Branches?
Abstract
This paper examines the effect of the internet on brick-and-mortar bank branches using a dynamic discrete game. The main feature of the model is that banks receive a chance to open or close a branch randomly in continuous time. A continuous time setting can accurately model the sequence of events where bank branches are open and closed sequentially during the year whereas discrete time setting assumes that every event occurs simultaneously at the beginning of the year. This is also supported by the established dates of bank branches, which are scattered throughout the year, provided in the Summary of Deposits data set by Federal Deposit Insurance Corporation (FDIC).The model consists of two stages: i) the first stage where banks choose the number of branches in continuous time; and ii) the second stage where consumers decide which bank to make their deposits and banks decide the deposit rate given the number of branches. The model is estimated using panel data including the number of bank branches, internet connections, and market and bank characteristics. The estimation starts from the second stage model using a nested logit model. Then, for the first stage, I apply the nested pseudo likelihood (NPL) estimator which is newly introduced in continuous time by my another ongoing research, Blevins and Kim (2019). The results imply that the internet crowds out bank branches in small markets. Counterfactual simulations from preliminary results show that increase in the internet connections accelerates closures in bank branches. I am now estimating the second stage model to examine how consumers were affected by the bank branch closures induced by the rise of the internet.
This paper contributes to literature in several ways. First, this adds a new insight on banks’ branching strategy as new technology is introduced and there exist only a handful of studies on the effect of the internet in banking industry. Second, although online banking is rapidly replacing bank branches, there still exist consumers who still prefer branches for their banking. Thus, it is crucial to estimate how the internet is affecting bank branches and consumer welfare. Third, I use a continuous time model which better approximates the reality. This avoids the model misspecification problem also presented in Blevins and Kim (2019) which show that estimating a dynamic discrete game in discrete time when the data is generated in continuous time causes a large bias.
Does Trading Spur Specialization? Evidence From Patenting
Abstract
Exploiting staggered establishments of patent exchanges in China, we examine how patent trading affects firm innovation and specialization. We find that patent trading and in-house innovation are complements for patent sellers, whereas they are substitutes for the buyers. Our findings demonstrate that the market for technology induces (i) specialization between patent buyers and sellers, (ii) specialization between patent licensors and licensees, and (iii) specialization based on a firm’s R&D efficiency. All these three patterns of specialization indicate that a firm’s response to an emerging market for technology hinges on its comparative advantages. Firms with a comparative advantage in creating innovation redirect their resources from advertising to patenting activities, whereas firms with a comparative advantage in commercializing innovation switch their effort from patenting to advertising activities. Moreover, enhanced patent trading contributes to improved firm performance and increasing market concentration.Our findings suggest patent trading promotes comparative-advantage-based specialization and enhances firm performance.Endogenous Timing in Equity Crowdfunding
Abstract
I propose a model in which backers decide pledging timing in equity crowdfunding platforms. When the cost of early pledging is sufficiently low, only backers with sufficiently high valuations promote the project to other backers by pledging early. Instead, when the cost is high enough, and hence no types of backers have incentive to pledge early, the entrepreneur herself promotes the project by lowering share prices. As a result, higher costs lead to lower asset prices. I show that when the cost is high enough, higher likelihood of early pledging decreases project success rate. Moreover, among projects with moderate valuations, project with a higher valuation has a lower success rate. I further show that higher costs can improve the expected quality of the funded projects.Exchange-Traded Funds and Municipal Bond Market
Abstract
This paper examines the effect of exchange-traded funds' (ETFs) information disclosure on the pricing and trading of municipal bonds. I find that a bond held by ETFs has significantly smaller price dispersion than a similar, non-ETF-held bond. The effect is stronger for retail trades than for institutional trades, suggesting ETF holding disclosure brings about higher benefit for investors who lack information ex-ante. ETF-held bonds are also more liquid and associated with higher trading activity. Similar effect is documented in the primary market, as when a municipality has outstanding bonds held by ETFs, its newly issued bonds have lower yield, higher price, and lower price dispersion. Overall, my findings indicate that ETF holding disclosure is an important channel through which investors can gain additional insight into the pricing of municipal bonds.Extracting Statistical Factors When Betas Are Time-Varying
Abstract
This paper deals with identification and inference on the unobservable conditional factor space and its dimension in large unbalanced panels of asset returns. The model specification is nonparametric regarding the way the loadings vary in time as functions of common shocks and individual characteristics. The number of active factors can also be time-varying as an effect of the changing macroeconomic environment. The method deploys Instrumental Variables (IV) which have full-rank covariation with the factor betas in the cross-section. It allows for a large dimension of the vector generating the conditioning information by machine learning techniques. In an empirical application, we infer the conditional factor space in the panel of monthly returns of individual stocks in the CRSP dataset between January 1971 and December 2017.Financial Statement Complexity and Syndicated Loan Structure
Abstract
We investigate the relationship between borrowers’ financial reporting complexity and their syndicated loan structure. By studying the interactions of two mechanisms of financial statement complexity (information content vs. management obfuscation), we predict and find that such complexity is positively associated with the likelihood of having a non-relationship lead arranger and negatively associated with the likelihood of having non-relationship participants. We also provide evidence that complexity is positively associated with the fraction of the deal retained by the lead arranger in a syndicated loan, as well as with the loan spread. Our findings highlight that information complexity can affect the important characteristics of the syndicated loan structure.Fintech Lending and Sales Manipulation
Abstract
Debt enforcement is a challenging issue. Fintech payments companies as lenders possess a potential solution to weak debt enforcement. Their location in the payments chain gives them the enforcement technology to automatically deduct repayment at source before borrowing merchants receive their sales proceeds. We use transaction data from an Indian Fintech payments processor, offering such loans with automatic, sales-linked repayment. We show that borrowing merchants discontinuously reduce electronic sales processed through the processor right after the loan disbursal, potentially diverting to other means of payments. Defaulting and delayed borrowers exhibit larger discontinuity, pointing towards a diversion of electronic sales to intentionally default. Borrowers with better credit scores show a higher incidence of manipulation, which we relate to their better outside opportunities. Using the 2018 cash crunch episode, we find evidence for merchants diverting electronic sales to cash. The results underscore how competition from cash and other lenders limit the effectiveness of this enforcement technology.Firm Life-Cycle in Mergers and Acquisitions
Abstract
This paper shows that target firm's life-cycle affects company's acquisition decision. I propose a novel measure for the firm life-cycle, based on the portfolio of company's products, and I find that transactions are more likely between firms in the same life-cycle. Moreover, deals in which the acquirer and the target belong to the same life-cycle yield higher stock returns and stronger future asset growth compared with the rest of the transactions. The analysis reveals that firm life-cycle synergies constitute a fundamental factor in merger and acquisition deals.Group Identity and Agency Frictions: Evidence Using Big Data
Abstract
This paper examines whether similarity in social identities between a manager and the board affects executive compensation, firm value, and hence, agency frictions. By using a novel dataset on surnames with multiple identities (native language, native place, and caste), developed by merging micro census data of 474 million Indians with Linguistic Survey of India data, I provide evidence that the firms with a shared group identity between a manager and the board do well compare to other firms and due to in-group favoritism, managers of such firms earn higher compensation. These results are stronger for group identity based on native language and native place. I also find that the firm benefits from taking on the cost of in-group favoritism as it reduces the agency frictions and increases firm value in the long run. These results are robust to the endogeneity test, managerial influence on firm, college ties, ties from past employment, and various other checks.Heterogeneity in CDS Coverage
Abstract
The total long-term liabilities combined for S&P 500 companies increased three-fold in the past 18 years from $1700 billion to $5300 billion. On the contrary the Credit Default Swaps (CDS) market did not follow this pattern and 30% of the S&P 500 companies never had CDS despite having long term debt. Also, while some large companies such as Bed Bath & Beyond, Moody's Corp, Netapp Inc. do not have CDS on their outstanding debt, their direct competitors do. As of now, the cause of this puzzling heterogeneity in CDS coverage remains unclear. In this paper we fill this gap by showing that the demand for CDS is causally related to the structure of bond ownership. In particular the number of investors holding the underlying bond (breadth) and the concentration of ownership (institutional depth) affect the demand for CDS. For any given issuer, the intersection between these two attributes changes the financial risk borne by investors determining an increased or reduced demand for protection. If bond ownership is atomistic and numerosity of investors is high, default risk is fragmented and diversified away reducing the demand for CDS. At the same time, diffused ownership can make the individual investors too small to influence the governance of bond issuing firm and thus, increases the demand for protection. We test the two rival hypotheses on a unique, partially hand-collected sample of CDS issued on S&P 500 companies over a 4 year period using over 180,000 bond ownership observations obtained from Lipper eMAXX. Controlling for a number of covariates the data supports the managerial influence hypothesis that high numerosity and atomicity increases the demand for CDS. Our results have important normative implications in the regulation of CDS markets and naked CDS strategies.Heterogeneous Asset Pricing Model Preferences by Investor Type: Evidence from Separate Accounts
Abstract
This paper examines the top drivers of investor flows into US Separate Account Composites and US Mutual Funds. I find that for both actively managed US Separate Account Composites and actively managed US Mutual Funds, Morningstar rating is the most significant predictor of flows and supersedes the CAPM model alpha, Fama-French 3 Factor model alpha, Fama-French-Carhart 4 Factor model alpha and other measures of risk-adjusted return. The Sharpe ratio and Information ratio, both widely-used fund performance measures, also do not predict flows as well as the Morningstar rating. Surprisingly, the aforementioned results appear to hold for passively managed funds as well. Within factor models, the CAPM explains mutual fund flows best and the 4 Factor model explains separate account composite flows best, indicating a potentially higher average degree of sophistication for separate account investors. I also find a slight increase in the sophistication level of factor models used in more recent time periods.Heterogeneous Investor Response to New Risks in Financial Markets: Evidence from Climate Change Litigation
Abstract
This paper studies how investors respond to an increase in environmental litigation risk following a new lawsuit. I look at both defendants and competitors to address the potential selection effect of litigation. I find a cumulative abnormal return of around -5% for both defendants and competitors, so it is unlikely investors are not monitoring this risk. I also find large investors increase holdings by about 100,000 shares in the quarter of a lawsuit so it is unlikely that asymmetric information among investors is a big issue for this risk. I also find ESG investors increase holdings by about 100,000 shares so it does not seem that investors are redistributing shares based on their preferences for environmental factors. Since large and ESG investors are increasing holdings, it seems more likely that investors redistribute based on a comparative advantage to engage with the firm and improve environmental factors. I find environmental shareholder proposals decrease by around 2% following litigation. I interpret this result that large investors may be better able to engage privately and there is a substitution from public to private engagement.Incentivization or Expropriation? All ESOPs Are Not Created Equal
Abstract
Why do Chinese listed firms adopt Employee Stock Ownership Plans (ESOPs)? Chinese listed firms with high salaries experience improvement in productivity and shareholder value, but those with high leverage, intercorporate loans, and separation of ownership and control have increased financial distress risk after ESOP adoptions. Firms with deteriorating performance are smaller and tend to announce ESOP adoptions when market sentiment is high. Their controlling shareholders tend to use earnings management, leveraged ESOPs, and ESOPs with high participation rates to inflate the stock prices and then cash out soon after ESOP adoption announcements, siphoning billions of RMB from minority investors.Industrial Policy and Asset Prices: Evidence from the Made in China 2025 Policy
Abstract
We study the link between industrial policy and asset prices by using the Made in China 2025 industrial policy, announced in May 2015, as an external shock. We track Chinese firms and U.S. firms in ten high-tech industries targeted by the policy. In the short run, stock prices, measured by cumulative abnormal returns (CARs), increase significantly for both Chinese and U.S. firms, by 9.9% and 2.8%, respectively. However, in the long run, Chinese firms’ CARs drop heavily, while U.S. firms’ CARs continually increase. Chinese firms do not receive additional state support after the policy assignment, nor do they respond to the policy by expanding investment or employment. In addition, Chinese firms’ profitability declines dramatically by on average of 91% after the policy announcement. We conclude that the policy only boosts market reaction in the short run, but does not promote targeted industries longer term.Industry Dynamics and Capital Structure (Non)Commitment
Abstract
We study the interaction between industry dynamics and equity holders’ inability to commit to future funding in a competitive equilibrium model. Equity holders are unable to commit to future debt level, but at the same time have to make entry, exit and production decisions in face of competition. The inability to commit significantly increases the cost of debt financing and hinders potential entries into the market. The resultant higher output price, nonetheless, alleviates the debt-equity conflict of the incumbents. More importantly, the leverage dynamics in conjunction with technology shock shapes the distribution of the firm universe, escalating the industry turnover and leverage. Our model also provides a micro-foundation for the highly positive skewed distribution of leverage ratio and empirically accounts for it. Overall, the model sheds new lights on real impacts of equity holders’ inability to commit to future debt level.Initial Endowments and Access to Finance: The Role of Neolithic Transformation
Abstract
We evaluate the influences of the timing of Neolithic transformation, when the hunter-gatherers became the first farmers, on contemporary financial development. Exploiting two world-wide survey datasets that containing 139,759 firms and 142,280 households, we find that early Neolithic transformation, as a natural endowment, has endured positive influence on access to finance today via shaping the quality of legal institutions. Earlier transition well predicts more developed financial systems, fewer obstacles that firms faced in raising capital and easier access to financial services among households. The positive effects hold when we use regional high-resolution measures, and when we instrument Neolithic timing using six biogeography characteristics. Combining pre-industrial institutions, we identify state capacity as enduring influencing channel, which provides modern finance with more regulated environment.Institutional Herding and Corporate Debt Issuance
Abstract
Is institutional herding in financial markets merely a sideshow, or does it affect corporate financing decisions? In this paper, we investigate the impact of mutual funds herding on corporate debt issuance. We find that herding is significantly related to the timing of bond issuance and the choice of bonds over loans. Specifically, firms are more likely to issue new bonds at the time of high levels of herding in their existing bonds. Furthermore, the firm's tendency to opt for bonds over loans increases with the level of herding. The impact of herding is concentrated on buy-herding rather than sell-herding, among firms with greater information asymmetry, and for times when the bond market is opaque. Our results are robust to a matching- and an instrumental variable approach. Moreover, firms enjoy lower offering yield spreads and bond prices become more informativeness following mutual funds herding. Overall, our findings are consistent with the positive view of herding that it enhances information environment of the firm.Intellectual Property Protection in M&A Negotiations
Abstract
In this paper, I show that a major share of the value of target firm’s intellectual property can be protected from expropriation by the acquirer through negotiating a compensating bidder termination fee (BTF), which is paid to the target in case the acquirer abandons the deal. I apply a capitalization model for intangible capital stocks to proxy for the component of intellectual property in target firm’s market value. The results suggest that, on average, for every dollar of target firm’s R&D capital stock, roughly 16 cents of protective share is incorporated in the BTF. I strengthen my causal interpretation with an instrument variables approach that exploits exogenous industry-level variation in R&D worker quota. The relation between target firm’s innovation activity and BTF size is more pronounced, if the target is a pioneer in its technology sector, if the target operates in an industry that sells unique products, if the target is assigned to the hightech or healthcare industry, and if the target mentions “trade secrets” in its 10-K report filed with the SEC prior to deal announcement. The effect is further increasing in the degree of technological proximity as well as product market rivalry between acquirer and target. Extending prior research at the intersection of innovation, law, and M&A, this paper concludes that BTFs serve as a contract mechanism that provide target firms compensation for revelation of sensitive information in M&A negotiations if acquirers terminate deals. The option to include BTFs in M&A contracts thereby increases acquirers’ incentives to close the deal and increases targets’ ex-ante incentives to reveal innovative secret information.Vertical Integration and Mortgage Risk
Abstract
Using acquisition events between broker/dealer banks and mortgage lenders as shocks to local mortgage markets, we find vertical integration led to riskier mortgage lending practices both ex ante and ex post. With individual mortgage level data, we find vertical integration led to one percent increase in average loan-to-value ratio, which translated to a five percent increase in household leverage, and a 211 percent increase in one-year foreclosure rate. Moreover, vertically integrated lenders originated more and loans with higher interest rates loans than other lenders. We also show the existence of spillover effects: local competitors underwrote riskier loans in response to competition. Finally, local exposure to vertical integration is associated with local house price run-up before 2007, and collapse afterwards.Is Mutual Fund Family Retirement Money Smart?
Abstract
Using data on investments of fund family employees in their 401(k) plans, I show that employee flows predict fund performance up to two years. The predictive power is stronger when fund family employees are located close to fund managers, pointing to employees exploiting their proximity to managers to learn about the managers’ skill. The results are not driven by plan design, portfolio managers’ ownership, or cross-subsidization. The top quintile of funds in terms of employee flows outperforms the bottom quintile by 1.6% annually in terms of Carhart Alpha, suggesting that other investors can benefit by mimicking fund employees.Is There a Trade-Off between Protecting Investors and Promoting Entrepreneurial Activity? Evidence from Angel Financing
Abstract
This paper studies how changes in investor protection regulations affect local entrepreneurial activity, relying on the heterogeneous impact of a 2011 SEC regulation change on the definition of accredited investors across U.S. cities. Using a difference-in-differences approach, I show that cities more affected by the regulation change experienced a significantly larger decrease in local angel financing, entrepreneurial activity, innovation output, employment, and sales. I find that small business loans and second-lien mortgages became entrepreneurs’ substitutes for angel investment. My cost-benefit analysis suggests that the costs of protecting angel investors through the 2011 regulation change outweigh its benefits.Know Thyself: Free Credit Reports and the Retail Mortgage Mortgage Market
Abstract
Under imprecise creditworthiness information, borrowers may take erroneous credit decisions. Credit reports---which record one’s creditworthiness---became free in entire US in 2005, while they already had been free in seven states. Exploiting this in a difference-in-differences setting, I show that cheaper credit reports to consumers improved mortgage market outcomes. It resulted in increased mortgage demand and approvals, more origination to creditworthy borrowers, and reduced defaults and subprime population fraction. Also, lenders’ financial performance improved, and more consumers became first-time homeowners. Additional findings, including increased interest rates, suggest a demand-driven channel in which borrowers learn their creditworthiness from credit reports.Labor Mobility and Capital Misallocation in the Mutual Fund Industry
Abstract
If capital won’t come to fund managers, fund managers will go to capital. I document that fund managers move across mutual fund firms to manage amounts of capital that better match their skill, which improves the allocative efficiency of capital across fund managers. For causal identification, I exploit exogenous shocks to fund managers’ ability to switch firms due to state-level changes to non-compete laws. In states that strengthen the enforceability of non-compete agreements, the propensity of fund managers to switch mutual fund firms is halved, capital misallocation across managers increases by about 10%, and the sum of monthly value added of managers declines by over $25 million. These results indicate that fund managers’ mobility across firms plays an important role in the efficient allocation of capital within the mutual fund industry.Language Skills and Stock Market Participation: Evidence from Immigrants
Abstract
Do language skills affect investment decisions? This paper addresses this question by identifying the effects of English proficiency on stock market participation of immigrants in the U.S. and Australia. To establish causality, we construct an instrumental variable for English proficiency by exploiting the phenomenon that younger children acquire languages more easily than older children. We find that better English language skills significantly increase the probability of stock market participation and the portfolio share of stocks. Moreover, we provide evidence that the language effect on stock market participation operates through the reduction in information costs and the increase in social trust.Machine Learning Classification Methods and Portfolio Allocation: An Examination of Market Efficiency
Abstract
We design a novel empirical framework to examine market efficiency through out-of-sample (OOS) predictability. We frame the classic empirical asset pricing problem as a machine learning classification problem. We construct classification models to predict return states. The prediction-based portfolios beat the market in time series and cross-sections with significant economic gains. We directly measure prediction accuracies. For each model, we introduce a novel application of binomial test to test the accuracy of 3.34 million return state predictions. Our models can generate information about the relation between future returns and historical information. The establishment of predictability questions the correctness of prices.Uncovering Sparsity and Heterogeneity in Firm-Level Return Predictability Using Machine Learning
Abstract
We develop an approach that combines the estimation of monthly firm-level expected returns with an assignment of firms to (possibly) latent groups, both based upon observable characteristics, using machine learning principles with linear models. The best performing methods are flexible two-stage sparse models that capture group-membership predictive relationships. Our results uncover sparsity together with firms' heterogeneity based on their characteristics that improve both predictability and interpretability. We propose statistical tests based on nonparametric bootstrapping for our results, and detail how different characteristics may matter for different groups of firms, making comparisons to the existing literature.Monetary Surprises, Debt Structure and Credit Misallocation
Abstract
How does firm dynamically adjust its capital and debt structure in response to interest rate risk? I document a new fact that, at the aggregate level, a tightening monetary policy is associated with a rise in bank debt and a decline in market debt, which is not driven by cyclicality. Using micro-data, I find that financially unconstrained firms substitute loans for corporate bonds while financially constrained firms have a higher probability of issuing new equity. I develop a dynamic, heterogeneous firm model to quantitatively explain these patterns. Loan (bond) is modeled as risk-free (defaultable) debt. Firms trade-off higher intermediation cost of loan against default risk of bond for the optimal debt structure. An unanticipated interest rate hike raises default risk (cost of market debt). Firms with default risk and unbinding collateral constraint substitute loan for bond and therefore credit is reallocated from constrained firms to unconstrained firms. This generates credit misallocation and it raises the dispersion of marginal product of capital. The model implies a debt structure channel of monetary policy on firm's investment, borrowing and valuation. The economic mechanism emphasizes that the firm's preserved debt financing flexibility is an important determinant of firms' adjustment in response to interest rate risk. The model solutions quantitatively match the data.Opioid Prescription Rates and Asset Prices—Assessment of Causal Effects
Abstract
We explore the link between county-level opioid prescription rates and asset prices, specifically, stock returns of firms headquartered in that county, as well as real estate prices. In order to establish the causal effects of opioid prescription rates on firm stock returns, we first apply an instrumental variable (IV) regression approach and use the number of clandestine drug laboratories in a county to be the instrumental variable. The results provide robust evidence that county-level opioid prescription rates have a negative causal effect on the equity returns of firms headquartered in that county. Furthermore, we analyze the effect of Medical Board of California's 2014 regulatory revision aimed at reducing controlled substance overdose due to prescriptions and implement a difference-in-differences (DiD) estimation. The DiD estimation results show that this policy change has a positive dynamic effect on Californian firms' equity returns. We also find that the opioid prescription reduction assistance program provided by California Health Care Foundation (CHCF) to certain counties in California helps to raise the median prices of existing single-family homes in those counties by $28,678 on average.Political Corruption and Firm Access to the Initial Public Offering Market
Abstract
The study examines the causal effect of political corruption on firms access to capital. Politically corrupt environment increases underpricing and thereby imposes costs on firms. The effect intensifies with increased percentage of a firm's operations concentrated around the headquarter locations. Underwriters play a vital role in promoting IPOs and lowering information asymmetry in a corrupt environment. Political corruption does not diminish the likelihood of pre-IPO shareholders' achieving wealth gains. Overall, empirical evidence supports the notion that political corruption causes business uncertainty and a high degree of information asymmetry in the market.Present Bias, Asset Allocation and the Yield Curve
Abstract
This paper presents a present-biased general equilibrium model that explains many features of bond behavior. Present-biased investors increase (decrease) short-term (long-term) hedge demands compared to standard preferences. Hence, present bias drives up (down) short-term bond prices (yields) and drives down (up) long-term bond prices (yields), explaining the bond premium puzzle. The model produces realistic bond behavior with a present-bias factor of beta=0.35 and a long-term annual discount factor of delta=0.97, in line with the experimental literature. Bond behavior is best explained for a present-bias interval of at most 1 year, providing an estimate for the investor's duration of the present.Pricing Climate Change Risk in Corporate Bonds
Abstract
Using a firm’s geographic footprint to measure its exposure to sea level rise (SLR), I find that corporate bonds bear a climate risk premium upon issuance. A one standard deviation increase in firms’ SLR exposure is associated with a 7 basis point premium, representing a 3% increase in average yield spread. This effect is more pronounced for geographically concentrated firms, and within industries vulnerable to extreme weather conditions. I do not find evidence that credit rating agencies account for SLR exposure at bond issuance. Results are robust to placebo tests and inverse propensity weighting to address possible endogeneity.Pricing Information: Experimental Evidence
Abstract
We study information pricing in a laboratory experiment where subjects predict the next observation of a random walk process. At each step they can buy a signal that may be useful for making predictions. Participants know the parameters of the process, but not the informativeness of the signal. We find that subjects correctly price precise and imprecise signals, but overprice signals of average precision. This pattern persists even when the true precision level is revealed. We show that when participants do not buy the signal, they display optimism and use momentum and reversal strategies in forecasts. Although buying the signals spares them from these biases, relatively more biased participants do not value the signals higher than their peers.Propagating Trade Protectionism
Abstract
This paper studies the importance of global supply chains for the propagation of trade policy shocks. Protectionist measures recently adopted by the U.S. had adverse economic consequences for industries throughout the world, including in particular targeted industries, but also domestic industries as well as third parties uninvolved in the trade conflicts. Global supply chain linkages between industries explain the propagation of concentrated trade actions. They reinforce the direct effects on targeted industries and give rise to spillover effects that are negative and larger than direct effects. These findings suggest that global supply chains raise the economic costs of interventionist trade policy.Reading Tea Leaves: Benefits and Costs of Learning about Future Demand Shocks
Abstract
This paper studies how learning about future demand shocks affects investor behavior and market outcomes. Under the noisy rational expectations equilibrium framework, I show that with extra access to information about future demand shocks, investors learn less about current dividend and demand shocks, and allocate more precision to signals about future demand shocks via the front-running channel and a new \textit{future uncertainty channel}. The future uncertainty channel works on two levels: on the aggregate level, market-wise dividend information acquisition decreases, which makes dividends riskier and leads to higher risk premia; on the individual level, information acquisition on future demand shocks decreases the conditional variance of future excess payoffs. Thus the investor's \textit{ex ante} utility improves. I also show that price volatility increases due to more heterogeneous private information, and that learning about future demand shocks makes price more informative in the future but less informative in the present. I discuss, in the end, how learning about future demand shocks affects the market in practice through the examples of index recomposition and the BlackRock-BGI acquisition case.Funding Stability and Bank Liquidity
Abstract
In this paper, I study how rollover funding risk affects banks’ asset illiquidity choice. I address this question using the 2016 US money market fund reform, which reduced the availability of unstable MMF funding and imposed regulations that mitigated the incentives of MMF investors to run. Using this regulatory shock and a sample of MMF-borrowing international banks, I find that a reduction in rollover risk following MMF reform leads to an increase in the fraction of lending on bank balance sheets. I supplement this analysis with detailed corporate loan-level data and find that the increase in lending is concentrated in the most illiquid loan categories. This evidence supports the theory that unstable funding discourages investment in illiquid assets by exposing banks to fire-sale risk.Safe Asset Carry Trade
Abstract
We provide the first systematic asset pricing analysis of one of the main categories of near-money or safe assets, the repurchase agreement (repo). Heterogeneity in repo rates allows for a remunerative carry trade. The return on this carry trade, our carry factor, together with a market factor explain the temporal and cross-sectional variation in repo rates within a no-arbitrage framework: While the market factor determines the level of short-term interest rates, the carry factor accounts for the cross-sectional dispersion. Consistent with the safe asset literature, the carry factor reflects heterogeneity in convenience premia and is explained by the safety premium, the liquidity premium, and the opportunity cost of holding money. Our carry factor helps explain the cross-section on the returns on bonds even after accounting for standard bond pricing factors and after controlling for measures of bond safety and liquidity.Separating Retail and Investment Banking: Evidence from the UK
Abstract
The idea of separating retail and investment banking remains controversial. Exploiting the introduction of UK ring-fencing requirements in 2019, we document novel implications of such separation for credit and liquidity supply, competition, and risk-taking via a funding structure channel. By preventing universal banks from using retail deposits to fund investment banking activities, this separation leads universal banks to rebalance their activities towards domestic mortgage lending and away from supplying credit lines and underwriting services to large corporates. By redirecting the benefits of deposit funding towards the retail market, this rebalancing reduces the cost of credit for households, without eroding lending standards. However the rebalancing also increases mortgage market concentration and risk-taking by smaller banks via indirect competition effects.Shaped by Public Attention: Gender Equality, Female Executives and Product Market Performance
Abstract
We analyze the effects of public attention to social issues on product market performance. Specifically, we focus on public attention to gender equality and find that firms with a greater industry-adjusted female executive presence in the top management team experience significantly stronger product market performance only if the public attention to gender equality is particularly high. We find evidence more consistent with a social identity channel that suggests external (pull) rather than internal (push) factors underlying our findings, which become more significant for firms with corporate principal customers. We use the #MeToo movement as an exogenous shock to public attention of gender equality and find consistent results. Our study provides new insights on the role of public attention in shaping the relation between corporate female-friendly culture and industry dynamics.Short-Termist CEO Compensation in Speculative Markets: A Controlled Experiment
Abstract
Bolton, Scheinkman, and Xiong (2006) analyze a setting where investors disagree and short-sale constraints cause pessimistic views of stock prices to be sidelined, which leads to speculative stock prices. A theoretical implication of the model is that existing shareholders can exploit the speculative stock prices by (1) designing managerial compensation contracts that encourage short-term performance and (2) subsequently selling their shares to more optimistic investors. We document empirical support for this theory by finding that an exogenous removal of short-sale constraints curbs the provision of short-term incentives, an effect reflected in longer CEO compensation duration. The effect is concentrated among stocks with high investor disagreement and short-term-oriented institutional ownership. Consistent with prior work, we also find that longer CEO compensation duration leads to longer CEO investment horizons, less over-investment, and less earnings management.Social Inflation
Abstract
Social inflation refers to steeply rising insurance rates due to social factors such as large jury awards and broader definitions of liability. This paper is the first to study the risk of social inflation and its economic consequences. Using a novel dataset that spans jury verdicts, financial statements, and rate filings for commercial auto liability insurance, I find that the number of verdicts and settlements exceeding $50 million has increased almost threefold from 2011 to 2019. To highlight the role of these developments in insurance pricing, I build a model of social inflation and show that social inflation risk has a “double kick” effect on insurance price through increased effective marginal cost and interaction with the capital requirement. I then estimate the causal impact of social inflation risk on insurance rates throught a triple-difference framework. Ultimately, I uncover an important new source of aggregate risk that affects the stability of the insurance sector and the economic activities that depend on it.Does Interaction on Social Media Drive Extremeness or Moderation?
Abstract
Using comment streams on Seeking Alpha articles, we examine whether interacting on social media increases or moderates the extremeness of investors’ opinions. Unlike findings from political science, we find that interaction moderates extremeness. Comments become less extreme over the sequence of comments for a given article, within individual comment sub-threads, and over a single user’s comments for a given article. Extremeness moderation occurs both following earnings announcements and during non-earnings-announcement windows. Extremeness reduction is stronger when the article users are commenting on is more moderate, and when more commenters are self-identified (i.e., not anonymous). Finally, results suggest that the extremeness reduction triggered by Seeking Alpha articles has market implications. Differences of opinion captured by stock-based measures decrease significantly after Seeking Alpha articles with comments, but not after analyst forecast days or high-news days. Our results provide the first evidence of the influence of social media interaction on the updating of individuals’ opinions.Speaking with One Voice: Shareholder Collaboration on Activism
Abstract
Governing a firm via "voice" requires sufficient shares to speak with one voice. However, the lead activist of a shareholder campaign typically only owns a small fraction of the target firm's shares. How can a small activist afford costly campaigns and exert considerable influence over a firm? By developing a game-theoretic model, we show that the lead activist with small stakes is able to govern the firm as a large shareholder through building a coalition with fellow active shareholders. We find how the activist communicates and collaborates with fellow active shareholders to build and sustain the coalition. Surprisingly, a smaller activist's coalition is more likely to win support from passive shareholders. The interplay between the collaboration with active shareholders and the support of passive shareholders endogenously determines the optimal size of the activist's coalition.Specialization Effect of Non-technological Strategic Alliances on Innovation
Abstract
We examine the spillover effect of non-technological strategic alliances, which constitutes majority of alliances, on innovation. Through using the combined reporting status of US states as an instrument variable for non-tech alliances, we find that non-tech alliances result in more innovation output as measured by the number of patents and citations. We also show that the non-tech alliance driven increase in the number of patents is due to exploitation activities (developing specialization), as reflected by increased patents in firm’s existing technological areas and more self-citations and backward citations. These findings imply that, even in the absence of exploration activities (knowledge acquisition), strategic alliances can still be beneficial for partner firms through spurring specialization.Externalities of the Sharing Economy: Evidence from Ridesharing and the Local Housing Market
Abstract
This study highlights the externalities of the sharing economy on local economies. Using the introduction of Uber X as a staggered shock, I assess how ridesharing influences the local housing market through the interaction with public transit. After ridesharing’s entry, housing prices and market rents increase at the zip code level. The effect is more pronounced in locations with greater access to public transit and lower driving probability, consistent with the notion that ridesharing complements public transit. Similarly, there is a larger increase in housing prices and rents in zip codes with larger populations, lower median ages and more minorities, consistent with Uber X users’ characteristics. Also, price appreciation is strongest for houses that are just beyond walking distance to public transit, suggesting that ridesharing helps solve the “last mile” problem and redistributes the public transit premium. Overall, this study highlights the externalities of the sharing economy and provides important policy implications.Splitting and Shuffling: Institutional Trading Motives and Order Submissions across Brokers
Abstract
This paper studies order submission strategies by institutional investors when trading on private information. By merging institutional daily transactions with original/confidential 13F filings, I separate informed trades from uninformed ones. Informed large orders tend to be split across more brokers and over more days. While same brokers tend to work uninformed large orders over multiple days, the brokers who facilitated early parts of broken-up informed orders rarely receive the remaining parts of the same orders on later days. Institutional investors also provide camouflage for their informed orders by mixing an informed order with other uninformed orders simultaneously sent to the same broker. As a result, a higher degree of shuffling a portfolio of orders is associated with a larger share of informed trading volume. The splitting and shuffling strategies designed to conceal informed trades from brokers and other market participants tend to lower institutional trading costs, especially on informed orders.Voluntary Disclosure, Price Informativeness, and Efficient Investment
Abstract
I analyze a manager's decision to disclose private information when the stock market is a source of information for corporate investment-making. A manager with long-term incentives discloses her private information only if it crowds-in informed trading and increases the manager's ability to learn from the market. However, this ex-post disclosing behavior results in two crowding-out effects: First, it crowds-out informed trading in situations where the manager withholds her private information. Second, voluntary disclosure results in an ex-ante average decline in price informativeness. Paradoxically, ex-post voluntary disclosure aimed at stimulating informed trading distorts the market's feedback-providing role and restrains efficient investment-making. Long-term incentives induce this disclosing behavior and thus cause a novel form of investment inefficiency.Capital Ratios and Systemic Risk: Regulating non-banks in the syndicated lending market
Abstract
Capital adequacy ratios are a popular way of adding a resiliency buffer to financial firms' balance sheets, but every firm active in the lending market may react differently to the tightening of capital constraints. When applied to financial institutions that are less risk-averse than banks, an increase in capital ratio requirements may increase the level of risk in the lending system, as some institutions prioritise a search-for-yield at the expense of risk management. Assuming a mean-variance preference structure on lenders, I calibrate risk-aversion parameters and funding costs to each lenders in the syndicated loan market and run two counterfactual experiments to assess policy effectiveness at tackling systemic risk. I show, using a structural model of syndicated lending, that regulating non-banks leads them to take on more risk, therefore increasing the overall level of risk in the system, even as the probability of default of the average borrower decreases, the tail risk increases. I also confirm that simply modifying capital ratio on banks leads to a decrease in systemic risk, as expected and as implemented in Basel III. This paper serves as a cautionary tale of pursuing blanket macro-prudential policies when the actors in the market have heterogeneous business models and preferences.Surfing the Cycle: Cyclical Investment Opportunities and Firms' Risky Financial Assets
Abstract
This paper studies why non-financial firms invest in risky financial assets. Within a dynamic corporate finance model with macroeconomic fluctuations, I show that firms can use risky financial assets to transfer liquidity from states with low aggregate investment opportunities to states with high aggregate investment opportunities. Specifically, when investment funding demand is pro-cyclical and external financing is costly, risky financial assets with pro-cyclical returns can increase firm value by improving the match between internal cash flows and investment opportunities. Therefore, investing in risky financial assets can be naturally optimal for the firm from a macro perspective. Based on U.S. firm data scraped from the SEC 10-K filings using a machine learning algorithm, I find empirical evidence consistent with this mechanism: (1) time-serially, the value of risky financial assets is positively correlated with the corporate investment rate; (2) cross-sectionally, firms with pro-cyclical investment funding demand in excess of profits hold more risky financial assets. The empirical results are robust to adding variables to control for the risk-seeking, poor corporate governance and CEO overconfidence channels.Taking No Chances: Lender Monitoring and Corporate Acquisitions
Abstract
Using mergers between firms’ existing lenders as shocks to monitoring incentives and bargaining power, I find that intensified lender monitoring significantly reduces borrowers’ public takeover activities. The effect is driven by mergers involving lead lenders, and becomes stronger for less bank-dependent firms with more risk-taking tendencies. Lender mergers reduce not only acquisitions that are value-destroying to shareholders but also value-enhancing ones. Deals that do happen on average create no additional shareholder value and target cash-rich firms with stable incomes. These results suggest that lender monitoring mitigates agency concerns, yet also leads to over-conservative firm behavior.Testing Disagreement Models
Abstract
We provide plausibly identified evidence for the role of investor disagreement in asset pricing. Our natural experiment exploits the staggered implementation of EDGAR, which induces a reduction in investor disagreement with no accompanying changes in company fundamentals, disclosure quality, or earnings management. The reduction in disagreement leads to lower stock price crash risk. The effect is more pronounced for stocks with binding short-sale constraints and high investor optimism. The reduction in disagreement also leads to higher subsequent returns. Our results provide evidence consistent with models of investor disagreement.Textual Ambiguity in Financial Disclosures and Information Asymmetry among Investors
Abstract
Prior literature documents a temporary increase in information asymmetry between sophisticated and unsophisticated traders around earnings announcements and management forecasts. This increase occurs because sophisticated traders can capitalize on released information relatively quickly. In this study, we find when management uses more ambiguous tone in 10-K and 10-Q filings, the resulting spike in information asymmetry around the filing window is significantly lower than for firms which use less ambiguous tone. This suggests sophisticated traders are less able to extract useful information from financial disclosures when management itself is less certain about the prospects of the firm. Similarly, when financial statements are less readable, the resulting spike in information asymmetry is also lower.Mispricing and Anomalies: An Exogenous Shock to Short Selling from the Dividend Tax Law Change
Abstract
We study the causal effect of short selling on asset pricing anomalies by exploiting a novel exogenous shock to short selling. After the Job and Growth Tax Relief Reconciliation Act (JGTRRA) of 2003, equity lenders are reluctant to lend shares around the dividend record dates because substitute dividends that they would receive are taxed at ordinary income rates while qualified dividends are taxed at 15 percent, thus creating a negative shock to short selling. Using arguably the most comprehensive set of anomalies to date and the difference-in-differences (DID) regression framework, we find that anomalies become stronger after the dividend record months in the post-JGTRRA periods, driven by stronger mispricing in the dividend record months. We further show that the effect mainly comes from the overpriced stocks. Overall, our results provide strong evidence that most anomalies are likely due to mispricing, with valuation anomalies as an exception.The Economics of Non-competition Clauses
Abstract
A crucial organizational decision is the degree of access an employee is provided to the critical resource of the firm. The key tradeoff is that higher degree of access increases the employee's productivity inside the firm, but also enables the agent to fiercely compete upon leaving the firm. We study how non-competition agreements shape the optimal contract between a firm and the agent, whose distinctive characteristic is ability. We find that the lowest ability agents are subject to the strongest restrictions, while agents close to the average ability of the firm have increasingly milder covenants. cmmnt{to facilitate their participation}However, for the high ability agents, the intensity of noncompete is not decreasing any further as these agents can make the most use of the access granted to them. In an extension, we consider monetary transfers (wages). The firm prefers to offer a restrictive noncompete and pay higher wages when the employee's leave can cause high damages. The socially optimal regulation of the intensity of noncompete takes into account that a strong noncompete allows employment for lower ability agents, resulting in larger firm size, but restricts mobility when it is socially beneficial.The Effect of Eliminating Cross-Shareholdings on Firm Value: Evidence from Korean Business Groups
Abstract
Cross-shareholdings and firm value is a challenging question due to endogeneity problems of research on ownership structure. I investigate market value effects of cross-shareholdings, exploiting a regulation in Korea that prohibits new cross-ownership of business groups over 5 trillion won in combined assets. I find its overall positive value effect on the affiliated firms of target business groups, but significant costs of eliminating pre-existent cross-shares. The costs are higher for affiliates with a greater disparity in cash-flow and voting rights, distant from controlling shareholders’ direct ownership, and highly dependent on the internal capital market. The results suggest that removing cross-shareholdings improve agency conflicts while imposing potential costs.The Game Changer: Regulatory Reform and Multiple Credit Ratings
Abstract
This paper examines the change in the regulatory use of multiple credit ratings after the Dodd-Frank Act (Dodd-Frank). We find that post Dodd-Frank reform firms are less likely to demand a third rating (typically from Fitch) to support their new corporate bond issues. The reduction in the demand for a third rating is more prevalent among firms with ratings near the high yield (HY) - investment grade (IG) boundary, particularly for firms with HY-rated bonds. Third ratings also become less informative post Dodd-Frank, with a much weaker market impact on credit spreads for firms with S&P and Moody’s ratings on opposite sides of the HY-IG boundary. We provide new evidence on the effect of Dodd-Frank in curbing corporate borrowers’ strategic use of multiple credit ratings and the direct implications for their increased cost of borrowing.The Impact of Product Market Characteristics on Firms’ Strategies in Patent Litigation
Abstract
We use a compound real options model to investigate firms' strategic interactions in intellectual property litigation. Subject to financing constraints, an alleged infringer firm (“challenger”) and an infringed firm (“incumbent”) pay for their ongoing litigation cost using operating cash flows from product market profits. We consider the challenger's strategy to exit the market during litigation due to shortage of funds, the incumbent's strategy to withdraw from value-reducing litigation or to force the challenger to exit the market by a threat to litigate, and firms' strategies to set up royalty payments to avoid a lawsuit, or to settle with each other after a lawsuit is filed. By focusing on each firm's ability and willingness to pay for litigation costs, we find that product market characteristics such as the challenger's profit relative to the incumbent's loss of profits due to the alleged infringement (“gain-to-loss ratio”) has to be high enough for settlements to be possible. Settlements are also more likely in less volatile product markets, with more questionable patent validity, and when litigation costs are similar for the two firms. Our model generates new testable implications regarding IP litigation with financing considerations.The Impact of Social Media on Venture Capital Financing: Evidence from Twitter Interactions
Abstract
We examine the impact of information acquisition through social media on venture capital (VC) investment structures. We find evidence that social media engagement volume can affect VC staging and syndication as well as the probability of successful exit. VC firms are more likely to reduce the frequency of stage financing and less likely to syndicate with each other when portfolio companies’ social media accounts are more active and have higher engagement volume with their followers. We collect a unique data set from Twitter API and examine the impact of owned social media (OSM) and earned social media (ESM) on VC investment structures. Overall, our results show that entrepreneurial firms with higher OSM and ESM engagement volume are expected to have fewer financing rounds, a smaller number of members in their VC syndicates, a lower probability of VC syndication, a higher probability of an IPO exit and higher funding across all rounds. Our findings are robust to a variety of alternative model specifications, subsamples, controlling for endogeneity in VC staging and syndication, selection biases and machine learning approaches.The Portfolio Composition Effect
Abstract
Does the evaluation of a portfolio of stocks depend on its composition of winner and loser stocks? To test this, we define a simple, counting-based measure of performance – the number of winner relative to the number of loser stocks in a portfolio – and examine how this composition measure affects individuals’ willingness to invest in a portfolio. We derive testable predictions for the proposed composition measure from a framework which combines category-based thinking with mental accounting. To test our predictions, we demonstrate an approach to arrive at more reliable results in empirical asset pricing: we combine well-identified experimental results on individual investment decisions with the analysis of historical stock market data. Consistent with our predictions, we find across all experiments that individuals allocate larger investments to portfolios with more winner than loser stocks relative to alternative portfolios with more loser than winner stocks, although both portfolios (1) have realized identical overall portfolio returns and (2) show identical expected risk-return characteristics. Building on our experimental findings, we analyze fund flows of exchange-traded funds on leading equity market indices. We identify that the proposed portfolio composition measure is positively related to future net fund flows.The Safety Demand in a Monetary Union
Abstract
This paper studies why a monetary union benefits from common fiscal policy in the context of a structural safety demand. National governments optimally choose a fiscal policy that results in the underproduction of public safety compared to the globally efficient allocation. Governments cannot commit to the efficient fiscal policy as they benefit from unilaterally lowering spending so that households acquire foreign bonds with no fiscal cost instead of relying on domestic bonds for safety. The underproduction of public safety is especially severe in a monetary union since foreign bonds are not exposed to exchange rate risk. In a monetary union the globally efficient allocation can be implemented by an increase in common public spending funded by a common safe asset. The common spending should not substitute national spending, and repaying the debt using common resources dominates transferring the loan in case one member state cannot repay its share. Yet, also without transfers or common resources a Pareto improvement can be achieved.The Technical Default Spread
Abstract
We build a dynamic model in which endogenous loan covenants allocate investment control rights between borrowers and lenders, and study its implications for investment, risk-taking, and asset prices in the cross-section. When borrowers enter technical default by breaching a covenant, control rights switch from borrowers to lenders. Lenders optimally choose low-risk projects, thus mitigating borrowers’ risk-shifting incentives and reducing the firm’s cost of equity. A calibrated version of our model allows us to match the technical default spread that we find in the data: firms that are closer to technical default earn on average 4% lower future returns than firms that are further away from their technical default thresholds. We argue theoretically and show empirically that the technical default spread arises from different economic forces than the distress anomaly.Trend Factor in China: The Role of Large Individual Trading
Abstract
We propose a 4-factor model for the Chinese stock market by adding a trend factor into the market, size, and value of Liu, Stambaugh, and Yuan’s (2019) 3-factor model. Because of up to 80% of individual trading, the trend factor captures salient relevant price and volume trends, and earns a monthly Sharpe ratio of 0.48, much greater than that of the market (0.11), size (0.20), and value (0.28). The 4-factor model explains well a number of stylized facts and anomalies of the Chinese stock market. It also explains well mutual fund returns, serving as an analogue of Carhart’s (1997) model in China.United States Populist Rhetoric and Currency Returns
Abstract
We develop a novel measure of U.S. populist rhetoric. Aggregate Populist Rhetoric (APR) Index spikes around populist events. We decompose the APR Index into sub-indices. We show that APR Index and International Relations sub-index are negatively priced in the cross-section of currency excess returns. Currencies that perform well (badly) when U.S. populist rhetoric is high yield low (high) expected excess returns. Investors require high risk premium for holding currencies which underperform in times of rising U.S. populist rhetoric, especially in the post-crisis period. A long-short strategy that buys (sells) currencies with high (low) exposure to U.S. populism offers diversification benefits.Volatility Uncertainty and Jumps
Abstract
This paper analyzes the joint dynamics of S&P 500 jumps and volatility using option-implied information. Our results indicate that volatility is not related to the evolution of jumps but the uncertainty about volatility is. More uncertainty about future volatility shifts the expected return distribution to the left, such that negative price jumps are more likely and positive price jumps are less likely. We highlight the unique information content in volatility uncertainty beyond other moments and further show that it significantly predicts realized price jumps. Our results have strong implications for structural option pricing models. Unlike commonly assumed, we cast doubt on the existence of a linear link between the arrival of jumps and volatility.When Financial Advice Leads to Positive Performance: Evidence from Repeated Client-Advisor Interactions
Abstract
Many investors are subject to mental accounting, leaning on past experiences when making decisions. Does the performance of the financial products they purchased - on their own or with the help of a financial advisor - influence the decisions they make in the future? To find out, I use a sample of 3,455 calls between German investors and their financial advisor between 2005 and 2015. If the advisor's recommendations led to a positive performance in the past, chances of the investor following financial advice again increases by 3.4%, whilst decreasing the probability of completely refraining from financial advice by 10%. It increases portfolio values and causes a transitory turnover boost. Positive performance also leads to higher risks and costs of recommended products in the future.Which Factors with Price-Impact Costs?
Abstract
We show that the squared Sharpe ratio criterion considered by Barillas and Shanken (2017) is no longer appropriate to compare factor models in the presence of price-impact costs. Instead, we propose comparing factor models in terms of their mean-variance utility net of price-impact costs and develop a formal statistical test to compare nested and non-nested factor models. Empirically, we find that price-impact costs change the relative performance of factor models. For instance, while in the absence of costs a seven-factor model considered in DeMiguel, Martin-Utrera, Nogales, and Uppal (2020) is the best low-dimensional model we consider, in the presence of price-impact costs the six-factor model of Fama and French (2018) is better.JEL Classifications
- G0 - General