The journal of finance , tập 66, số 2, 2011 4

Vol 66 April 2011 No Editor Co-Editor CAMPBELL R HARVEY Duke University JOHN GRAHAM Duke University Associate Editors VIRAL ACHARYA New York University FRANCIS A LONGSTAFF University of California, Los Angeles ANAT R ADMATI Stanford University HANNO LUSTIG University of California, Los Angeles ANDREW ANG Columbia University ANDREW METRICK Yale University KERRY BACK Rice University TOBIAS J MOSKOWITZ University of Chicago MALCOLM BAKER Harvard University DAVID K MUSTO University of Pennsylvania NICHOLAS C BARBERIS Yale University STEFAN NAGEL Stanford University NITTAI K BERGMAN Massachusetts Institute of Technology TERRANCE ODEAN University of California, Berkeley HENDRIK BESSEMBINDER University of Utah CHRISTINE A PARLOUR University of California, Berkeley MICHAEL W BRANDT Duke University ALON BRAV Duke University MARKUS K BRUNNERMEIER Princeton University DAVID A CHAPMAN Boston College MIKHAIL CHERNOV London School of Economics JENNIFER S CONRAD University of North Carolina FRANCESCA CORNELLI London Business School BERNARD DUMAS INSEAD DAVID HIRSHLEIFER University of California, Irvine BURTON HOLLIFIELD Carnegie Mellon University HARRISON HONG Princeton University NARASIMHAN JEGADEESH Emory University WEI JIANG Columbia University STEVEN N KAPLAN University of Chicago JONATHAN M KARPOFF University of Washington ARVIND KRISHNAMURTHY Northwestern University MICHAEL LEMMON University of Utah ´ L˘ UBOS˘ PASTOR University of Chicago LASSE H PEDERSEN New York University MITCHELL A PETERSEN Northwestern University MANJU PURI Duke University RAGHURAM RAJAN University of Chicago JOSHUA RAUH Northwestern University MICHAEL R ROBERTS University of Pennsylvania ANTOINETTE SCHOAR Massachusetts Institute of Technology HENRI SERVAES London Business School ANIL SHIVDASANI University of North Carolina RICHARD STANTON University of California, Berkeley ANNETTE VISSING-JORGENSEN Northwestern University ANDREW WINTON University of Minnesota Business Manager DAVID H PYLE University of California, Berkeley Assistant Editor WENDY WASHBURN HELP DESK The Latest Information Our World Wide Web Site For the latest information about the journal, about our annual meeting, or about other announcements of interest to our membership, consult our web site at http://www.afajof.org Subscription Questions or Problems Address Changes Journal Customer Services: For ordering information, claims, and any enquiry concerning your journal subscription, please go to interscience.wiley.com/support or contact your nearest office Americas: Email: cs-journals@wiley.com; Tel: +1 781 388 8598 or +1 800 835 6770 (toll free in the USA & Canada) Europe, Middle East and Africa: Email: cs-journals@wiley.com; Tel: +44 (0) 1865 778315 Asia Pacific: Email: cs-journals@wiley.com; Tel: +65 6511 8000 Japan: For Japanese speaking support, email: cs-japan@wiley.com; Tel: +65 6511 8010 or Tel (toll-free): 005 316 50 480 Further Japanese customer support is also available at www.interscience.wiley.com/support Visit our Online Customer Self-Help available in six languages at www.interscience wiley.com/support Permissions to Reprint Materials from the Journal of Finance C 2011 The American Finance Association All rights reserved With the exception of fair dealing for the purposes of research or private study, or criticism or review, no part of this publication may be reproduced, stored or transmitted in any form or by any means without the prior permission in writing from the copyright holder Authorization to photocopy items for internal and personal use is granted by the copyright holder for libraries and other users of the Copyright Clearance Center (CCC), 222 Rosewood Drive, Danvers, MA 01923, USA (www.copyright.com), provided the appropriate fee is paid directly to the CCC This consent does not extend to other kinds of copying, such as copying for general distribution for advertising or promotional purposes, for creating new collective works, or for resale For information regarding reproduction for classroom use, please see the AFA policy statement in the back of this issue Advertising in the Journal For advertising information, please visit the journal’s web site at www.afajof.org or contact the Academic and Science, Advertising Sales Coordinator, at corporatesalesusa@wiley.com 350 Main St Malden, MA 02148 Phone: 781.388.8532, Fax: 781.338.8532 Association Business Those having business with the American Finance Association, or members wishing to volunteer their service or ideas that the association might develop, should contact the Executive Secretary and Treasurer: Prof David Pyle, American Finance Association, University of California, Berkeley—Haas School of Business, 545 Student Services Building, Berkeley, CA 94720-1900 Telephone and Fax: (510) 642-2397; email: pyle@haas.berkeley.edu Volume 66 CONTENTS for APRIL 2011 No ANNOUNCEMENT OF 2010 SMITH BREEDEN AND BRATTLE GROUP PRIZES v ARTICLES Bankruptcy and the Collateral Channel EFRAIM BENMELECH and NITTAI K BERGMAN 337 Public Information and Coordination: Evidence from a Credit Registry Expansion ANDREW HERTZBERG, JOSE´ MAR´IA LIBERTI, and DANIEL PARAVISINI Security Issue Timing: What Do Managers Know, and When Do They Know It? DIRK JENTER, KATHARINA LEWELLEN, and JEROLD B WARNER Private Equity and Long-Run Investment: The Case of Innovation ¨ JOSH LERNER, MORTEN SORENSEN, and PER STROMBERG Do Buyouts (Still) Create Value? SHOURUN GUO, EDITH S HOTCHKISS, and WEIHONG SONG 379 413 445 479 The Joy of Giving or Assisted Living? Using Strategic Surveys to Separate Public Care Aversion from Bequest Motives JOHN AMERIKS, ANDREW CAPLIN, STEVEN LAUFER, and STIJN VAN NIEUWERBURGH 519 Corporate Governance, Product Market Competition, and Equity Prices XAVIER GIROUD and HOLGER M MUELLER 563 The Interim Trading Skills of Institutional Investors ANDY PUCKETT and XUEMIN (STERLING) YAN 601 Institutional Trade Persistence and Long-Term Equity Returns AMIL DASGUPTA, ANDREA PRAT, and MICHELA VERARDO 635 Local Dividend Clienteles BO BECKER, ZORAN IVKOVIC´ , and SCOTT WEISBENNER 655 MISCELLANEA 685 THE JOURNAL OF FINANCE • VOL LXVI, NO • APRIL 2011 SMITH BREEDEN PRIZES FOR 2010 First Prize Paper Joao F Gomes and Lukas Schmid Levered Returns April 2010 Distinguished Papers Joel Peress Product Market Competition, Insider Trading, and Stock Market Efficiency February 2010 Lauren Cohen, Andrea Frazzini, and Christopher Malloy Sell-Side School Ties August 2010 ¨ Richard C Green, Dan Li, and Norman Schurhoff Price Discovery in Illiquid Markets: Do Financial Asset Prices Rise Faster Than They Fall? October 2010 BRATTLE GROUP PRIZES FOR 2010 First Prize Paper Andrew Hertzberg, Jos´e M Liberti, and Daniel Paravisini Information and Incentives Inside the Firm: Evidence from Loan Officer Rotation June 2010 Distinguished Papers Thorsten Beck, Ross Levine, and Alexey Levkov Big Bad Banks? The Winners and Losers from Bank Deregulation in the United States October 2010 Jos´e M Liberti and Atif R Mian Collateral Spread and Financial Development February 2010 THE JOURNAL OF FINANCE • VOL LXVI, NO • APRIL 2011 Bankruptcy and the Collateral Channel EFRAIM BENMELECH and NITTAI K BERGMAN∗ ABSTRACT Do bankrupt firms impose negative externalities on their nonbankrupt competitors? We propose and analyze a collateral channel in which a firm’s bankruptcy reduces the collateral value of other industry participants, thereby increasing their cost of debt financing We identify the collateral channel using novel data of secured debt tranches issued by U.S airlines that include detailed descriptions of the underlying collateral pools Our estimates suggest that industry bankruptcies have a sizeable impact on the cost of debt financing of other industry participants We discuss how the collateral channel may lead to contagion effects that amplify the business cycle during industry downturns DO BANKRUPT FIRMS affect their solvent nonbankrupt competitors? Although a large body of research studies the consequences of bankruptcy reorganizations and liquidations for those firms that actually file for court protection (e.g., Asquith, Gertner, and Scharfstein (1994), Hotchkiss (1995), and Str¨omberg (2000)), little is known about the externalities that bankrupt firms impose on other firms operating in the same industry Any such externalities would be of particular concern, as they may give rise to self-reinforcing feedback loops that amplify the business cycle during industry downturns Indeed, the potential for contagion effects was of particular concern during the financial panic of 2007 to 2009, where insolvent bank liquidations and asset sell offs imposed “fire-sale” externalities on the economy at large (see, e.g., Kashyap, Rajan, and Stein (2008)) In this paper, we identify one channel through which bankrupt firms impose negative externalities on nonbankrupt competitors, namely, through their impact on collateral values We use the term “collateral channel” to describe this effect According to the collateral channel, one firm’s bankruptcy reduces the ∗ Benmelech is from Harvard University (the Department of Economics) and NBER, and Bergman is from MIT Sloan School of Management and NBER We thank Paul Asquith, Douglas Baird, John Campbell, John Cochrane, Lauren Cohen, Shawn Cole, Joshua Coval, Sergei Davydenko, Douglas Diamond, Luigi Guiso, Campbell Harvey (the Editor), Oliver Hart, John Heaton, Christian Leuz, Andrei Shleifer, Jeremy Stein, Heather Tookes, Aleh Tsyvinsky, Jeffrey Zwiebel, an associate editor of the Journal of Finance, two anonymous referees, and seminar participants at DePaul University, The Einaudi Institute for Economics and Finance in Rome, Harvard University, University of Chicago Booth School of Business, and the 2008 Financial Research Association Meeting Benmelech is grateful for financial support from the National Science Foundation under CAREER award SES-0847392 We also thank Robert Grundy and Phil Shewring from Airclaims, Inc Ricardo Enriquez and Apurv Jain provided excellent research assistance All errors are our own 337 338 The Journal of Finance R collateral values of other industry participants, particularly when the market for assets is relatively illiquid Because collateral plays an important role in raising debt finance, this reduction in collateral values increases the cost and reduces the availability of external finance across the entire industry Theory provides two interrelated reasons for the prediction that the bankruptcy of one industry participant lowers the collateral values of other industry participants First, a firm’s bankruptcy will increase the likelihood of asset sales and hence will place downward pressure on the value of similar assets, particularly when there are frictions in the secondary market For example, in an illiquid market, bankruptcy-induced sales of assets will create a disparity of supply over demand, causing asset prices to decline, at least temporarily (for evidence on asset fire sales, see Pulvino (1998, 1999)).1 In the context of real estate markets, whose collapse was of crucial importance in instigating and magnifying the crisis, Campbell, Giglio, and Pathak (2009) provide evidence of spillover effects in which house foreclosures reduce the price of other houses located in the same area The second reason that bankruptcies will tend to reduce collateral values is related to their impact on the demand for assets When a firm is in financial distress, its demand for industry assets will likely diminish, as the firm does not have and cannot easily raise the funding required to purchase industry assets (Shleifer and Vishny (1992), Kiyotaki and Moore (1997)) Thus, bankruptcies and financial distress reduce the demand for industry assets, again placing downward pressure on the value of collateral Reductions in demand for assets driven by bankruptcies and financial distress are currently evident in the difficulties the Federal Deposit Insurance Corporation is encountering in selling failed banks These difficulties have arisen because traditional buyers of failed banks—namely, other banks—are financially weak.2 Thus, due both to increased supply and reduced demand for industry assets, the collateral channel implies that bankruptcies increase the likelihood of asset fire sales, reducing collateral values industry wide This weakens the balance sheet of nonbankrupt firms, thereby raising their cost of debt capital Empirically, a number of important outcomes have been shown to be sensitive to the announcement of the bankruptcy of industry competitors For example, Lang and Stulz (1992) show that when a firm declares bankruptcy, on average, competitor firm stock prices react negatively Likewise, Hertzel and Officer (2008) and Jorion and Zhang (2007) examine the effect of bankruptcy on competitors’ loan yields and CDS spreads.3 Further support for fire sales is provided by Acharya, Bharath, and Srinivasan (2007), who show that recovery rates are lower when an industry is in distress Indeed, to partially solve this problem, the FDIC is looking outside the traditional market, at private equity funds, to infuse fresh capital into the banking system and purchase failed financial institutions See “New Rules Restrict Bank Sales,” New York Times, August 26, 2009 In related literature, Chevalier and Scharfstein (1995, 1996) and Phillips (1995) examine a contagion effect from firms in financial distress to other industry participants through a product market channel while Peek and Rosengren (1997, 2000) and Gan (2007a, 2007b) analyze a lending channel contagion effect from banks in distress to their corporate borrowers Bankruptcy and the Collateral Channel 339 However, identifying a causal link from the financial distress or bankruptcy filings of some players in an industry to the cost of capital of these firms’ solvent nonbankrupt competitors is difficult because bankruptcy filings and financial distress are potentially correlated with the state of the industry Financial distress and bankruptcy filings themselves thus convey industry-specific information, explaining, for example, negative industry-wide stock price reactions and loan pricing effects The question therefore remains: bankrupt firms affect their competitors in a causal manner or the observed adverse effects merely reflect changes in the economic environment faced by the industry at large? Using a novel data set of secured debt tranches issued by U.S airlines, we provide empirical support for the collateral channel Airlines in the United States issue tranches of secured debt known as equipment trust certificates (ETCs), enhanced equipment trust certificates (EETCs), and pass-through certificates (PTCs) We construct a sample of aircraft tranche issues and then obtain the serial number of all aircraft that were pledged as collateral For each of the debt tranches in our sample, we can identify precisely its underlying collateral We then identify the “collateral channel” off of both the time-series variation in bankruptcy filings by airlines, and the cross-sectional variation in the overlap between the aircraft types used as collateral for a specific debt tranche and the aircraft types operated by bankrupt airlines The richness of our data, which includes detailed information on tranches’ underlying collateral and airlines’ fleets, combined with the fairly large number of airline bankruptcies in our sample period, allows us to identify strategic externalities that are likely driven by a collateral channel rather than by an industry shock to the economic environment At heart, our identification strategy relies on analyzing the differential impact of an airline’s bankruptcy on the credit spread of tranches that are secured by aircraft of different model types According to the collateral channel hypothesis, tranches whose underlying collateral comprises model types that have a large amount of overlap with the fleet of the bankrupt airline should exhibit larger price declines than tranches whose collateral has little overlap with the bankrupt airline’s fleet For each tranche in our sample, we construct two measures of bankruptcyinduced collateral shocks The first measure tracks the evolution over time of the number of airlines in bankruptcy operating aircraft of the same model types as those serving as collateral for the tranche Because airlines tend to acquire aircraft of the same model types that they already operate, an increase in the first measure is associated with a reduction in the number of potential buyers of the underlying tranche collateral The second measure of collateral shocks tracks the number of aircraft operated by bankrupt airlines of the same model type as those serving as tranche collateral An increase in this second measure is associated with a greater supply of aircraft on the market that are similar to those serving as tranche collateral Increases in either of these two measures therefore tend to decrease the value of tranche collateral and hence increase credit spreads 340 The Journal of Finance R Using both measures, we find that bankruptcy-induced collateral shocks are indeed associated with higher tranche spreads For example, our univariate tests show that the mean spread of tranches with no potential buyers in bankruptcy is 208 basis points, while the mean spread of tranches with at least one potential buyer in bankruptcy is 339 basis points Moreover, our regression analysis shows that the results are robust to a battery of airline and tranche controls, as well as airline, tranche, and year fixed effects Our identification strategy allows us to identify only price reactions to bankruptcies of other industry participants However, through its influence on firms’ cost of capital, these price effects potentially have real effects such as reducing firms’ debt capacity and investment We further show that the effect of collateral shocks is temporary and confined to the duration of firm bankruptcies The temporary nature of the negative externality is consistent with price pressure effects driven by the collateral channel Still, given the long periods over which firms remain in bankruptcy, this temporary effect is sizeable Further, because bankruptcies are more prevalent during industry downturns, the bankruptcy-induced collateral channel—while temporary—has the potential to amplify the downturn of the industry We continue by showing that the effect of bankruptcy-induced collateral shocks on credit spreads is higher for less senior tranches with higher loan-tovalue (LTV) This is to be expected, as more junior tranches are more exposed to drops in the value of the underlying collateralized assets upon default Next, we analyze the interaction between the collateral channel and airline financial health Because airlines in poor financial health are more likely to default, the spread of these tranches should be more sensitive to underlying tranche liquidation values Measuring financial health using either airline profitability or a model of airline predicted probability of default, we find that the effect of collateral shocks on tranche spreads is more pronounced in high LTV tranches of airlines in poor financial health Finally, we analyze the interaction between collateral shocks and the redeployability of tranche underlying collateral and find that the positive relation between the number of potential buyers of tranche collateral that are in bankruptcy and tranche credit spreads is lower for tranches with more redeployable collateral Using a host of robustness tests and analysis, we show that our results are not driven by underlying industry conditions or by other forms of potential contagion unrelated to the collateral channel For example, we show that our results are not likely driven by sales pressure stemming from binding balance sheet constraints of ETC and EETC security holders, nor are they likely driven by reverse causality in which adverse shocks to the productivity of certain aircraft results in the bankruptcies of those airlines using these aircraft as well as an increase in the cost of capital for other users of these aircraft Further, our results are not driven by the provision of credit enhancement in the form of a liquidity facility The rest of the paper is organized as follows Section I provides the theoretical framework for the analysis and explains our identification methodology Bankruptcy and the Collateral Channel 341 Section II provides institutional details on the market for ETCs and EETCs Section III describes our data and the empirical measures Section IV presents the empirical analysis of the relation between bankruptcy-induced collateral shocks and credit spreads Section V concludes I Identification Strategy To analyze the collateral channel we focus on a single industry—airlines— and employ a unique identification strategy This strategy involves using information on collateral characteristics, collateral pricing, and the timing of airline bankruptcies in the following manner Airlines in the United States issue tranches of secured debt to finance their operations The debt is secured by a pool of aircraft serving as collateral Using filing prospectuses, we identify the model type of all aircraft that serve as collateral in each pool For each tranche, we obtain a time series of prices and obtain the dates and durations of all bankruptcy filings of airlines in the United States during the years 1994– 2007 In essence, our identification strategy consists of analyzing the differential impact of an airline’s bankruptcy on the price of tranches that are secured by aircraft of different model types The collateral channel hypothesis predicts that tranches whose underlying collateral comprises model types that have a large degree of overlap with the fleet of the bankrupt airline should exhibit larger price declines than tranches whose collateral has little overlap with the bankrupt airline’s fleet As explained above, an airline’s bankruptcy and the increased likelihood of the sale of part or all of the airline’s fleet will place downward pressure on the value of aircraft of the same model type Furthermore, as in Shleifer and Vishny (1992), because demand for a given aircraft model type stems to a large extent from airlines that already operate that model type, an airline’s financial distress and bankruptcy will reduce demand for the types of aircraft that it operates in its fleet For these two reasons—both increased supply of aircraft in the used market and reduced demand for certain aircraft—tranches secured by aircraft of model types exhibiting larger overlaps with the model types of the bankrupt airline’s fleet should experience larger price declines By using variation in the fleets of airlines going bankrupt and their degree of overlap with the type of aircraft serving as collateral for secured debt of other airlines, we can thus identify a collateral channel through which one firm’s bankruptcy affects other firms in the same industry Because we rely on the differential impact of bankruptcy on the credit spreads of tranches secured by aircraft of different model types within an airline, this identification strategy alleviates concerns that the results are driven by an information channel effect in which bankruptcies convey negative information common to all firms in the industry Moreover, we test our evidence for the collateral channel against alternative contagion-based explanations For example, we show that our results are not driven by contagion through credit enhancers or through holders of tranche securities 342 The Journal of Finance R In the next section, we describe in further detail the debt instruments used by airlines to issue secured debt and their development over time II Airline Equipment Trust Certificates ETCs and EETCs are aircraft asset-backed securities (ABS) that have been used since the early 1990s to finance the acquisitions of new aircraft.4 Aircraft ABSs are subject to Section 1110 protection, which provides relief from the automatic stay of assets in bankruptcy to creditors holding a secured interest in aircraft, strengthening the creditor rights of the holders of these securities The U.S Bankruptcy Code began to treat aircraft financing favorably in 1957, but it was not until 1979 that Congress amended the Bankruptcy Code and introduced Section 1110 protection, which provides creditors relief from the automatic stay On October 22, 1994, the Bankruptcy Code was further amended, and the rights of creditors under Section 1110 were strengthened The changes in the Bankruptcy Code increased the protection that Section 1110 provided to secured creditors and reduced the potential threat of legal challenge to secured aircraft This legal innovation affected the practice of secured lending in the airline industry The market for ETCs expanded and new financial innovations such as EETCs soon became the leading source of external financing of aircraft The amendments to Section 1110 led Moody’s to revise its ratings criteria such that securities that were issued after the enactment date received a rating up to two notches above issuing airlines’ senior unsecured rating In a traditional ETC, a trustee issues ETCs to investors and uses the proceeds to buy the aircraft, which is then leased to the airline Lease payments are then used to pay principal and interest on the certificates The collateral of ETCs typically includes only one or two aircraft For example, on August 24, 1990, American Airlines issued an ETC (1990 ETCs, Series P) maturing on March 4, 2014 The certificates were issued to finance approximately 77% of the equipment cost of one Boeing 757-223 (serial number 24583) passenger aircraft, including engines (Rolls-Royce RB211-535E4B) The proceeds from the ETC issue were $35.5 million, with a serial interest rate of 10.36% and a credit rating of A (S&P) and A1 (Moody’s) Increasing issuance costs led to the development of PTCs, which pooled a number of ETCs into a single security that was then backed by a pool of aircraft rather than just a single one Although PTCs increased diversification and reduced exposure to a single aircraft, the airline industry downturn in the early 1990s led to downgrades of many ETCs and PTCs to below investment grade and subsequently to a narrowed investor base During the mid-1990s, ETCs and PTCs were further modified into EETCs— which soon became the leading source of external finance of aircraft EETC Our discussion here draws heavily from Littlejohns and McGairl (1998), Morrell (2001), and Benmelech and Bergman (2009), who provide an extensive description of the market for airline ETCs and its historical evolution 676 The Journal of Finance R us to test further whether the change in dividend policy differs across mover and non-mover companies (also controlling for the effect of changes in firmspecific characteristics on both movers and non-movers) These specifications thus include the indicator variable Firm Moved (equal to one if the company’s headquarters moved in year t, and zero otherwise) The findings offer additional reassurance—by itself, the fact that the company headquarter locations have changed (controlling for everything else, of course) plays little role in the change in dividend policy.16 Moreover, the size and significance level of the coefficients associated with Local Seniors are highly comparable across the two analyses Naturally, the small sample size of the set of movers calls for caution in interpreting these results and, thus, in extrapolating these results to the general population of firms Nonetheless, the direction, magnitude, and significance of these results all line up in a manner consistent with the dividend demand hypothesis and our results presented elsewhere in the paper, thus offering complementary evidence based on an entirely different identification strategy III Potential Benefits of Satisfying Local Dividend Demand Having found effects consistent with the dividend demand hypothesis, we proceed with the third and final tier of our identification strategy In this section, we explore why managers might wish to respond to local seniors’ demand for dividends,17 whether there are benefits to such demand-induced payouts, as well as the mechanisms through which individual investor demand may affect corporate policy We consider two possible channels and offer suggestive evidence At the outset, we remark that the channels we discuss in this section are not mutually exclusive Moreover, none of these channels require that managers be explicitly informed about local retail investors’ age, or that they should feel goodwill toward local investors in general or local seniors in particular A Lower Turnover Graham and Kumar (2006) report that seniors are more likely to buy dividend-paying stocks in the weeks leading up to the ex-dividend day, and are more likely to buy stocks after they start to pay dividends We use the same brokerage data as in Graham and Kumar (2006) and build upon their result that seniors buy stocks after a company initiates dividends (or just before a company is to pay dividends) by studying individuals’ propensity to sell a stock 16 Coefficients associated with changes in the firm-level variables in these change regressions line up as expected See the Internet Appendix for the coefficient estimates associated with the other independent variables 17 Corporations generally are known to care, at least to some extent, about retail investors According to Brav et al (2005), executives believe that attracting retail investors to purchase their stock is an important motivation behind company dividend policy Local Dividend Clienteles 677 they hold already (i.e., what local seniors after they buy the stock) The brokerage data are extremely well suited for this purpose—to study the sale decision of a given investor and a given stock holding That is, the many observations of potential sale behavior across many investors enable us to obtain strong and robust results regarding holding periods We can thus ascertain, with a high degree of precision, what happens once investors purchase stock— given their characteristics (in this case geographic location and age), whether they are more likely, relative to others, to keep on holding the stock We test whether, conditional on owning the stock, local seniors have a substantially longer holding period than other types of investors This lower turnover may be attractive to company management, and a way to attract such loyal investors to own the stock in the first place is to pay dividends To test the “lower turnover” channel, we first conduct a completely nonparametric analysis in which we estimate the cumulative likelihood of sale of a given stock holding for four investor groups (comprising just over 30,000 households): potential sales by all individuals, potential sales by seniors (65 years of age or older), potential sales by local investors (the distance between the household and corporate headquarters is 250 miles or fewer—the local-distance metric used in Ivkovi´c and Weisbenner (2005)), and potential sales by local senior investors Tallying the sales made in a given month since purchase relative to the total number of potential sales at the beginning of the month produces nonparametric hazard rates for each month (i.e., the probability of selling in that month conditional on not having sold the stock up to that point), and cumulating those monthly hazard rates yields the cumulative probability distribution of sale as a function of time since purchase The four cumulative probability distributions are presented in Figure Whereas the median holding period for all investors (the first sample) is 15 months, the median holding period for local senior investors (the fourth sample) is substantially longer, 37 months This analysis, though compelling, does not take into consideration other potential motivations for sale, including stock performance since purchase, nor potential selection issues based on preferences for holding different kinds of stocks (each of which, by themselves, could lead to different selling patterns across individuals) Therefore, we also conduct a more stringent analysis by incorporating several covariates that capture investor sensitivity to past performance (be it for behavioral reasons or tax motivations) Our econometric framework for these analyses follows Ivkovi´c, Poterba, and Weisbenner (2005) closely We use the Cox proportional hazards model, which employs nonparametric estimates of baseline monthly hazards (that is, the probabilities of selling the stock in month t after purchase, conditional on no prior sale) The results of the Cox proportional hazards model, presented in the Internet Appendix, confirm the differences in the likelihood of selling stock displayed in Figure According to the estimates from the more rigorous proportional hazards model analyses, during a month, a local senior investor is 28% to 32% less likely to sell a share of stock than is a nonlocal nonsenior investor Aggregated over time, this difference results in much longer holding 678 The Journal of Finance R periods and much lower turnover for the group of investors who likely are attracted to purchase the stock of firms that pay dividends B Price Channel: Ex-Dividend Day Reaction Our second line of inquiry considers valuation effects Our identification strategy studies stock price movements around the ex-dividend day.18 We relate ex-dividend day returns to Local Seniors, a technique used previously to infer marginal tax rates (see, e.g., Elton and Gruber (1970), Perez-Gonzales (2003), and Graham and Kumar (2006)), as well as dividend demand valuation effects for nontax reasons (Graham and Kumar (2006)) Elton and Gruber (1970) divide ex-dividend price drops by the amount of the dividend and report that this ratio averages 0.78 Depending on the subpopulation of corporations they consider, Graham and Kumar (2006) report that this ratio in their study averages from 0.67 to 0.79 The price drop is less than the dividend paid, presumably because, for many investors, dividends face higher taxes than capital gains As Elton and Gruber (1970) point out, in a rational market, the price drop when the stock goes ex-dividend reflects the relative value of dividends and capital gains to the marginal stockholders Thus, a company whose owners face a lower dividend tax rate (or a higher tax rate on capital gains) should experience a bigger drop in the share price when the stock goes ex-dividend We apply a similar logic to the firms facing investors with a dividend preference: the ex-dividend day price drop, as a fraction of the dividend amount, should be large when demand for dividends is high As before, we use Local Seniors to proxy for dividend demand Alternative explanations for our findings on payout policy (e.g., that Local Seniors proxies for local economic conditions that lead to dividend supply by firms) not predict such a relation between ex-dividend day price behavior and Local Seniors We adjust the ex-dividend methodology of Elton and Gruber (1970) along two dimensions First, we use price changes from market close on the last cum-dividend day to the opening trade the following, ex-dividend day, and thereby focus more narrowly on the price change related to the loss of the dividend right This mitigates the risk of confounding the inference with the price changes taking place during the following, ex-dividend day Second, whereas the original Elton and Gruber (1970) methodology normalizes the price change by the dividend amount, our dependent variable is the relative price drop— the negative of the price change from the close of the last cum-dividend day to the open of the ex-dividend day divided by the closing price on the last cum-dividend day.19 Accordingly, our independent variables are the dividend amount scaled by the closing price, Local Seniors, and the interaction between the two (we also include Median Income, the median income of the county in which the company is headquartered and its interaction with the amount of 18 We thank Denis Gromb for suggesting this approach This methodology avoids normalizing by any small quantity (such as the dividend amount) and, therefore, is more statistically robust 19 Local Dividend Clienteles 679 the dividend).20 We estimate these regressions over the sample of stock returns surrounding ex-dividend days in the period from 1992 to 2007 Local Seniors and Median Income in off-Census years are estimated by linear interpolation (and extrapolation) of the figures from the Census years Table VII reports the results The first column provides a simple gauge; it close (abbreviated in the table as Div/P for readability) The features only Divi,t /Pi,t−1 regression coefficient estimate is 0.73, suggesting that, absent further controls, a one-dollar dividend is associated with a 73-cent price drop on the ex-dividend day, a figure that is very similar to the baseline figures from Elton and Gruber (1970) and Graham and Kumar (2006) close and its interaction with the indicator The second column features Divi,t /Pi,t−1 variable Small Firm (defined as in Table V), thereby allowing for differential valuations of dividends across the shareholders of small and large corporations The results match intuition and extant findings very well For large companies, the price drop on the ex-dividend day associated with a one-dollar dividend is 86 cents, whereas for smaller companies, more likely to be held by individuals and thus more likely to be subject to higher marginal tax rates, the price drop is 60 cents (the difference between the two is 26 cents, statistically significant at the 1% level) In the third column, we present evidence that the ex-dividend day price drop, that is, the valuation of dividends, varies across communities We add to the specification Local Seniors, the interaction Div/P × Local Seniors, Median Income, and the interaction Div/P × Median Income In light of our primary interest in how community characteristics affect the relation between the ex-dividend day price drop and the amount of the dividend (that is, the interactions of Local Seniors and Median Income with Div/P), the coefficients associated with Local Seniors and Median Income themselves are suppressed from the table for readability.21 Median Income, measured at the county level, is a geographically based measure of income intended to capture tax-related motivations from potential local shareholders Consistent with seniors’ demand for dividends, the price drop is positively related to Local Seniors (the coefficient estimate is 3.0, statistically significant at the 10% level), that is, the ex-dividend day price drop is larger in areas with a higher fraction of seniors The magnitude of this effect is substantial—a one-standard deviation increase in Local Seniors is associated with an increase in the ex-dividend day price drop as a fraction of the dividend of 0.09, or one-eighth of the average price drop from column (1) The relation between the ex-dividend day price drop and the interaction Div/P × Median Income is negative and statistically significant, consistent with smaller ex-dividend day price drops in regions in which income and, 20 The Elton and Gruber (1970) estimation corresponds to estimating this regression and weighting observations by the inverse of their dividend yield (that is, putting the most weight on observations pertaining to stocks with the lowest yields) Thus, our methodology generalizes the original approach from Elton and Gruber (1970) 21 The coefficient associated with Local Seniors in the specification presented in column (3) is 0.004 (SE = 0.005) and the coefficient associated with Median Income is 0.009 (SE = 0.004) 680 The Journal of Finance R Table VII Ex-dividend Day Returns This table presents results of estimating OLS regressions that relate the price drop on the exdividend date to Local Seniors and other covariates The estimation is carried out over the sample of stock returns surrounding ex-dividend days in the period from 1992 to 2007 The dependent variable is the relative price drop—the negative of the price change from the close of the last cum-dividend day to the open of the ex-dividend day divided by the closing price of the last cumdividend day The first column relates the price drop to the amount of the dividend scaled by the close closing price (Divi,t /Pi,t−1 , abbreviated in the table as Div/P for readability) In the second column, Div/P is interacted with the indicator variable Small Firm (defined as in Table V) to allow for differential effects across large and small firms In columns (3) and (4), two county-level variables are added to the specification: Local Seniors (the fraction of residents who are 65 years old or older in the county in which a firm is headquartered, as reported by the U.S Census Bureau) and Median Income (measured at the county level; this geographically based measure of income is intended to capture local tax-related motivations more directly) The third column includes Div/P, Local Seniors, Median Income, as well as interactions between Div/P and Local Seniors and Median Income, respectively The fourth column further allows for interactions of all independent variables featured in the third column with the indicator variable Small Firm Local Seniors and Median Income in off-Census years are estimated by linear interpolation (or extrapolation) of the figures from the Census years In light of the primary interest in how community characteristics affect the relation between the ex-dividend day price drop and the amount of the dividend, the coefficients associated with Local Seniors and Median Income, and their interactions with the Small Firm indicator variable are suppressed from the table for readability These coefficients are reported in the text of Section III.B Standard errors (shown in parentheses) allow for heteroskedasticity and are clustered by firm ∗∗∗ , ∗∗ , ∗ denote significance at the 1%, 5%, and 10% levels, respectively Div/P (1) (2) (3) (4) 0.73∗∗∗ (0.05) 0.86∗∗∗ (0.02) −0.26∗∗∗ (0.09) 0.60∗∗∗ (0.09) 0.36 (0.23) 0.92∗∗∗ (0.08) −0.89∗∗∗ (0.34) 0.04 (0.33) −0.19 (0.54) 4.9∗ (2.6) 4.7∗ (2.5) −0.01 (0.01) −0.10∗ (0.06) −0.11∗∗ (0.06) (Div/P) × Small Firm Total effect of Div/P for small firms (Div/P) × Local Seniors 3.0∗ (1.7) (Div/P) × Local Seniors × Small Firm Total effect of (Div/P) × Local Seniors for small firms −0.08∗∗ (0.03) (Div/P) × Median Income (Div/P) × Median Income × Small Firm Total effect of (Div/P) × Med Income for small firms Adjusted R2 Number of observations 0.196 116,933 0.201 116,933 0.198 116,370 0.205 116,370 presumably, tax rates are higher This result is consistent with marginal investors of at least some corporations being local individuals (and, thus, whether a firm is located in a community in which individuals have high income and high tax rates affects the valuation of the firm’s dividends on the ex-dividend day) However, given the unit of measurement of Median Income (in units of Local Dividend Clienteles 681 $100,000), its economic importance is very modest (a one-standard-deviation increase in the median income of the county, $15,500 in our sample, is associated with an ex-dividend day price drop as a fraction of the dividend that is only 0.01 smaller) Finally, in the fourth column all the variables from the third column are interacted with the indicator variable Small Firm (defined as in Table V) to allow for differential effects across small and large companies.22 Our prediction is that the sensitivity of the valuation of dividends to Local Seniors should be stronger for small firms than for large firms This is exactly what we find As shown in column (4), there is no relation between the ex-dividend day price drop and the presence of local seniors in the community (or the median income of the county) for large corporations Adding the coefficients on Div/P × Local Seniors and Div/P × Local Seniors × Small Firm provides the total effect of Local Seniors on the ex-dividend day price drop for small stocks Compared to the estimate across all stocks of 3.0 from column (2), the effect for small stocks is larger, 4.7 (statistically significant at the 10% level; the associated p-value is 0.06) Moreover, for small stocks a one-standard-deviation increase in Local Seniors is associated with an increase in the ex-dividend day price drop as a fraction of the dividend of 0.15, one-quarter of the average price drop from column (2) For small companies, which are more likely to rely on local shareholders, the ex-dividend price drop is also related to the median income of the county in which the company is headquartered in the direction predicted, but the economic magnitude of this local income effect again is very small In sum, these results extend the findings from Graham and Kumar (2006), who show that the price fall on the ex-dividend day is positively related to the age of the shareholders of the firm and negatively related to the income of the shareholders of the firm Our ex-dividend results are consistent with the notion that companies located in counties with higher fractions of seniors face stronger demand for dividends, and that such demand is associated with larger stock-price drops on ex-dividend days, but only for small companies, which are more likely to be affected by this geographically varying demand for dividends from the local senior population The valuation effects detected on the ex-dividend day and the prior evidence concerning lower turnover, while each suggestive on their own, together reinforce the dividend demand hypothesis Moreover, neither channel requires that managers be explicitly informed about local retail investors’ age IV Conclusion Miller and Modigliani (1961) raise the question of whether firms set policies and investors sort accordingly, or whether companies respond to the prefer22 Once again, for readability, in Table VII we not report the coefficients associated with Local Seniors and Median Income, nor we report their interactions with Small Firm, the small-firm indicator variable The coefficient associated with Local Seniors in the specification presented in column (4) is −0.001 (SE = 0.003) and the interaction of Local Seniors and Small Firm is –0.001 (SE = 0.011) The coefficient associated with Median Income is 0.001 (SE = 0.001) and the interaction of Median Income with Small Firm is 0.016 (SE = 0.070) 682 The Journal of Finance R ences of their current shareholders In this paper, we provide evidence consistent with the latter Specifically, we test for the effect of dividend demand on payout policy The tendency of older investors to hold dividend-paying stocks in combination with individual investors’ inclination to hold local stocks results in stronger dividend demand for companies located in areas with many seniors Demographics thus provide an empirical proxy for dividend demand, which we exploit in this paper to examine the broader question of whether the preferences of current owners influence corporate actions As predicted, we find a significant positive effect of Local Seniors, the fraction of seniors in the county in which a firm is located, on the firm’s propensity to pay dividends, its propensity to initiate dividends, and its dividend yield The effect of Local Seniors on the corporate decision to start paying dividends is particularly strong, of the same economic magnitude as other key determinants such as company size and age Because demographics are only a rough proxy for demand, our results can be interpreted as placing a lower bound on the effect of investor preferences on payout policy If there are other components of demand, the total effect of investor preferences on corporate policies may be larger Our results are robust to various methodologies and identification strategies, and are not supportive of alternative explanations (e.g., that firms located in areas with many seniors have low growth opportunities and, therefore, are more likely to pay out cash to shareholders) The main implication of our findings is that, at least to some extent, corporations respond to the preferences of their owners when setting payout policy We confirm that age determines dividend demand, consistent with the hypothesis of Miller and Modigliani (1961) and the evidence presented in Graham and Kumar (2006) We further show that there are dividend clienteles that vary geographically, creating differences in demand for dividends across firms that help explain some of the substantial cross-sectional heterogeneity as to why some companies pay dividends while others not Our findings thus suggest that there is important geographical variation in the financial conditions that firms face REFERENCES Allen, Franklin, Antonio E Bernardo, and Ivo Welch, 2000, A theory of dividends based on tax clienteles, Journal of Finance 55, 2499–2536 Baker, Malcolm, and Jeffrey Wurgler, 2004a, A catering theory of dividends, Journal of Finance 59, 1125–1165 Baker, Malcolm, and Jeffrey Wurgler, 2004b, Appearing and disappearing dividends: The link to catering incentives, Journal of Financial Economics 73, 271–288 Barber, Brad, and Terrance Odean, 2000, Trading is hazardous to your wealth: The common stock investment performance of individual investors, Journal of Finance 55, 773–806 Becker, Bo, 2007, Geographical segmentation of U.S capital markets, Journal of Financial Economics 85, 151–178 Brav, Alon, John R Graham, Campbell R Harvey, and Roni Michaely, 2005, Payout policy in the 21st century, Journal of Financial Economics 77, 483–527 Local Dividend Clienteles 683 Brown, Jeffrey, Nellie Liang, and Scott Weisbenner, 2007, Executive financial incentives and payout policy: Firm responses to the 2003 dividend tax cut, Journal of Finance 62, 1935–1965 Coval, Joshua D., and Tobias J Moskowitz, 1999, Home bias at home: Local equity preference in domestic portfolios, Journal of Finance 54, 1–39 Coval, Joshua D., and Tobias J Moskowitz, 2001, The geography of investment: Informed trading and asset prices, Journal of Political Economy 109, 811–841 Elton, Edward J., and Martin J Gruber, 1970, Marginal stockholder clientele effect, Review of Economics and Statistics 52, 68–74 Fama, Eugene F., and Kenneth R French, 2001, Disappearing dividends: Changing firm characteristics or lower propensity to pay? Journal of Financial Economics 60, 3–43 Graham, John R., 2003, Taxes and corporate finance: A review, Review of Financial Studies 16, 1074–1128 Graham, John R., and Alok Kumar, 2006, Do dividend clienteles exist? Evidence on dividend preferences of retail investors, Journal of Finance 59, 1125–1165 Grinblatt, Mark, and Matti Keloharju, 2001, How distance, language, and culture influence stockholdings and trades, Journal of Finance 56, 1053–1073 Guiso, Luigi, Paola Sapienza, and Luigi Zingales, 2004, Does local financial development matter? Quarterly Journal of Economics 119, 929–969 Hong, Harrison, Jeffrey Kubik, and Jeremy Stein, 2008, The only game in town: Stock-price consequences of local bias, Journal of Financial Economics 90, 20–37 Huberman, Gur, 2001, Familiarity breeds investment, Review of Financial Studies 14, 659–680 Ivkovi´c, Zoran, James Poterba, and Scott Weisbenner, 2005, Tax-motivated trading by individual investors, American Economic Review 95, 1605–1630 Ivkovi´c, Zoran, and Scott Weisbenner, 2005, Local does as local is: Information content of the geography of individual investors’ common stock investments, Journal of Finance 60, 267–306 Jayaratne, Jith, and Philip Strahan, 1996, The finance-growth Nexus: Evidence from bank branch deregulation, Quarterly Journal of Economics 111, 639–670 Lintner, John, 1956, Distribution of incomes of corporations among dividends, retained earnings, and taxes, American Economic Review 46, 97–113 Loughran, Tim, and Paul Schulz, 2005, Liquidity: Urban versus rural firms, Journal of Financial Economics 78, 341–374 Massa, Massimo, and Andrei Simonov, 2006, Hedging, familiarity and portfolio choice, Review of Financial Studies 19, 633–685 Miller, Merton H., and Franco Modigliani, 1961, Dividend policy, growth, and the valuation of shares, Journal of Business 34, 411–433 Perez-Gonzales, Francisco, 2003, Large shareholders and dividends: Evidence from U.S tax reforms, Working paper, Stanford University Peterson, Mitchell A., and Raghuram G Rajan, 2002, Does distance still matter? The information revolution in small business lending, Journal of Finance 57, 2533–2570 Pirinsky, Christo, and Quinghai Wang, 2006, Does corporate headquarters location matter for stock returns? Journal of Finance 61, 1991–2015 Scholz, John K., 1992, A direct examination of the dividend clientele hypothesis, Journal of Public Economics 49, 261–285 Shefrin, Hersh, and Meir Statman, 1984, Explaining investor preference for cash dividends, Journal of Financial Economics 13, 253–282 Shefrin, Hersh, and Richard H Thaler, 1988, The behavioral life-cycle hypothesis, Economic Inquiry 26, 609–643 Thaler, Richard H., and Hersh M Shefrin, 1981, An economic theory of self-control, Journal of Political Economy 89, 392–406 THE JOURNAL OF FINANCE r VOL LXVI, NO r APRIL 2011 MISCELLANEA The following articles have been accepted for publication in The Journal of Finance and are scheduled to appear in the June 2011 issue You can read the full text of all upcoming articles on the AFA website at the following address: http://www.afajof.org/journal/forthcoming.asp ARTICLES Viral V Acharya, Stewart C Myers, and Raghuram G Rajan, “The Internal Governance of Firms,” New York University, Massachusetts Institute of Technology, and University of Chicago Francis A Longstaff, “Municipal Debt and Marginal Tax Rates: Is There a Tax Premium in Asset Prices?” University of California, Los Angeles Dhammika Dharmapala, C Fritz Foley, and Kristin J Forbes, “Watch What I Do, Not What I Say: The Unintended Consequences of the Homeland Investment Act,” University of Illinois at Urbana-Champaign, Harvard Business School, and Massachusetts Institute of Technology Lorenzo Garlappi and Hong Yan, “Financial Distress and the Cross-Section of Equity Returns,” University of British Columbia and University of South Carolina Ronald W Masulis and Shawn Mobbs, “Are All Inside Directors the Same? Evidence from the External Directorship Market?” University of New South Wales and University of Alabama Stefan Nagel and Kenneth J Singleton, “Estimation and Evaluation of Conditional Asset Pricing Models,” Stanford University Jack Bao, Jun Pan, and Jiang Wang, “The Illiquidity of Corporate Bonds,” Ohio State University, Massachusetts Institute of Technology, and Massachusetts Institute of Technology Neal M Stoughton, Youchang Wu, and Josef Zechner, “Intermediated Investment Management,” UNSW Sydney, University of Wisconsin-Madison, and Vienna University of Economics and Business 685 686 The Journal of Finance R Ilona Babenko, Michael Lemmon, and Yuri Tserlukevich, “Employee Stock Options and Investment,” Arizona State University, University of Utah, and Arizona State University Trond M Døskeland and Hans K Hvide, “Do Individual Investors Have Asymmetric Information Based on Work Experience?” Norwegian School of Economics and Business Administration and University of Aberdeen THE JOURNAL OF FINANCE • VOL LXVI, NO • APRIL 2011 ANNOUNCEMENTS Annual Meeting: The Seventy Second Annual Meeting will be held in Chicago, Il, January 6–8, 2012 with Sheridan Titman as Program Chair Submissions closed March 11, 2011 2011 Fischer Black Prize: The American Finance Association is pleased to announce the award of the 2011 Fischer Black Prize to Xavier Gabaix, the Martin J Gruber Professor of Finance at the New York University Stern School of Business Professor Gabaix has many notable and highly original research contributions on a number of subjects in financial economics, including the level of compensation of corporate executives, and behaviorally influenced decision making and its influence on asset market behavior In some of his work, he has cleverly exploited axiom-based models of the shapes of the tails of probability distributions A hallmark of Professor Gabaix’s research style is his propensity to take unexpected directions 2011 AFA Fellow: The Fellows of the AFA have elected Professor Milton Harris, the Chicago Board of Trade Professor of Finance and Economics of the Booth School of Business of the University of Chicago as the 2011 Fellow Professor Harris has produced path breaking research in several areas of corporate finance and on the economic theory of information Among many important results, he is particularly recognized for his work on the design of securities, of firms, and more generally of organizations Professor Harris has frequently uncovered important new insights stemming from the role of asymmetric information, particular in settings involving contract design and agency relationships 2010 Election: The results of the 2010 election are: President: Raghuram Rajan, President-Elect: Sheridan Titman, Vice President Robert Stambaugh, Director Lasse Pedersen, Paola Sapienza, Raman Uppal Worldwide Directory of Finance Faculty: The AFA and the Department of Finance at Ohio State University have entered into a joint venture to maintain and enhance the finance faculty directory held on the OSU web site At present, information on over 3,000 finance professors and professionals is available in the directory An effort is being made to include all AFA members on this list and members are encouraged to provide information to the directory manager A link to the directory is available on the homepage or you can go directly to http://www.cob.ohio-state.edu/fin/findir/ Other Announcements Please go to our web site, www.afajof.org, for announcements regarding meetings, conferences, and research support 687 AMERICAN FINANCE ASSOCIATION Publisher of the Journal of Finance Prof David H Pyle Executive Secretary and Treasurer February 2011 To Those Seeking Permissions for Academic Classroom Use: Permission is granted to reproduce articles for classroom use by accredited, notfor-profit colleges and universities or their appointed agents without charge for: r classes of a faculty member who is a subscriber to The Journal of Finance r classes at a college or university with a library subscription to The Journal of Finance Articles also may be distributed for classroom use in electronic (pdf) form if they are stored on a password-protected web site at said institution or its agent Non-subscribers seeking to reproduce articles should contact Wiley-Blackwell Publishing Company (jrights@wiley.com) regarding permission This form is valid through February 1, 2012 University of California Berkeley Haas School of Business 545 Student Services Building Berkeley, CA 94720-1900 phone & fax: (510) 642-2397 STYLE INSTRUCTIONS (1)—-All submitted manuscripts must be original work that is not under submission at another journal or under consideration for publication in another form, such as a monograph or chapter of a book Authors of submitted papers are obligated not to submit their paper for publication elsewhere until an editorial decision is rendered on their submission Further, authors of accepted papers are prohibited from publishing the results in other publications that appear before the paper is published in the Journal, unless they receive approval for doing so from the managing editor (2)—-Authors must submit papers electronically Instructions for submission and manuscript preparation are available at the submission site, http://www.afajof.org/journal/submission.asp Manuscripts must be clearly typed with double spacing The pitch must not exceed 12 characters per inch, and the character height must be at least 10 points (3)—-The cover page shall contain the title of the paper and an abstract of not more than 100 words The title page should NOT include the names of the authors, their affiliations, or any other identifying information That information must be input separately as part of the on-line submission The abstract must also be entered or pasted into a separate text box as part of the on-line submission (4)—-An initial acknowledgement footnote should NOT be included Acknowledgements must be entered or pasted into a separate text box as part of the on-line submission (5)—-The introductory section must have no heading or number Subsequent headings should be given Roman numerals Subsection headings should be lettered A, B, C, etc (6)—-The article should end with a non-technical summary statement of the main conclusions Lengthy mathematical proofs and very extensive detailed tables should be placed in an appendix or omitted entirely The author should make every effort to explain the meaning of mathematical proofs Final versions of the manuscript should be prepared in accordance with the following additional guidelines: (7)—Footnotes Footnotes in the text must be numbered consecutively and typed on a separate page, double-spaced, following the reference section Footnotes to tables must also be double-spaced and typed on the bottom of the page with the table (8)—Tables Tables must be numbered with Roman numerals Please check that your text contains a reference to each table Indicate with a notation inserted in the text approximately where each table should be placed Type each table on a separate page at the end of the paper Tables must be self-contained, in the sense that the reader must be able to understand them without going back to the text of the paper Each table must have a title followed by a descriptive legend Authors must check tables to be sure that the title, column headings, captions, etc are clear and to the point (9)—Figures Figures must be numbered with Arabic numerals All figure captions must be typed in double space on a separate sheet following the footnotes A figure’s title should be part of the caption Figures must be self-contained Each figure must have a title followed by a descriptive legend Final figures for accepted papers must be submitted in native electronic form and uploaded as separate files on the submission site (10)—Equations All but very short mathematical expressions should be displayed on a separate line and centered Equations must be numbered consecutively on the right margin, using Arabic numerals in parentheses Use Greek letters only when necessary Do not use a dot over a variable to denote time derivative; only D operator notations are acceptable (11)—References References must be typed on a separate page, double-spaced, at the end of the paper References to publications in the text should appear as follows: “Jensen and Meckling (1976) report that ” At the end of the manuscript (before tables and figures), the complete list of references should be listed as follows: For monographs: Fama, Eugene F., and Merton H Miller, 1972, The Theory of Finance (Dryden Press, Hinsdale, III.) For contributions to collective works: Grossman, Sanford J., and Oliver D Hart, 1982, Corporate financial structure and managerial incentives, in John J McCall, ed.: The Economics of Information and Uncertainty (University of Chicago Press, Chicago, III.) For periodicals: Jensen, Michael C., and William H Meckling, 1976, Theory of the firm: Managerial behavior, agency costs and ownership structure, Journal of Financial Economics 3, 305–360 [...]... 0.0 0.0 5.0 311.0 0.077 0. 348 0. 648 0.872 1.171 1 .48 3 0.8 64 0.5 14 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.0 1.0 3.0 3.0 4. 0 5.0 4. 0 2.0 4, 8 14 3 ,4 21 3,0 56 2,9 37 2 ,4 97 1,8 26 2,8 34 1,0 03 1.859 3.817 39.208 61.598 70.8 84 83.939 54. 236 34. 126 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 57 17 2 74 273 282 311 2 64 181 4, 8 14 3 ,4 21 3,0 56 2,9 37 2 ,4 97 1,8 26 2,8 34 1,0 03 Panel A: Bankrupt Buyers and Number of Aircraft in Bankruptcy... 48 .316 (31.2 94) ≥0.5 45 .45 2∗∗ (80.668) 78.510 (108.399) −172.159 (186 .49 5) 42 .083 (32 .47 5) 32. 543 (23.992) 0.266 (4. 223) 50.2 54 (76.971) 144 .177∗∗∗ ( 34. 441 ) −7 54. 646 ∗∗∗ (218.162) 42 6.302∗∗∗ (1 34. 1 84) 705.267∗∗∗ ( 247 .669) 1,1 09.505∗∗ 49 7.782) 571.176 ( 1,6 60.159) 1,8 82.593 ( 1,3 37.861) 191.1 94 ∗∗ (44 .997) 46 .201 (46 .5 34) −611.588∗∗∗ (169.520) 33 .43 4 (122.091) 123.501 (128.668) 1 84. 908∗∗∗ (40 .49 1) − 1 ,4 28.775∗∗∗... 41 .212 (101.099) 30. 548 ( 34. 520) 3.506 (5.3 84) 30. 844 (179.205) 14. 459 (1 64. 320) − 1,5 21.139∗∗ ( 642 .168) 268.310 (47 7.368) Airbus and Boeing Boeing only Tranche Spread Tranche Spread 103 .49 9∗∗∗ (36.759) −38.579 (81.353 112.696∗∗∗ (41 .1 24) 77.172 (97.071) 61.196 (40 .156) 7. 948 (7.833) 122. 048 ( 94. 711) 146 .193∗∗∗ (46 .055) − 2,1 49 . 84 ∗∗ (41 7.980) −187 .42 0 (185 .46 3) −18.773 (39. 344 ) 0. 946 (7.012) −28.9 24. .. (116.073) 217.955∗∗ (91 .49 0) − 1,8 46 .818∗∗ (703.2 54) −2 14. 891 (41 8.2 64) Tranche Spread Tranche Spread 123.376∗∗∗ (42 .46 8) 17.332 (69.988) 26.121∗∗∗ (8.7 64) −27.966 (70 .41 6) 35.010 (71.103) −22.516 (13. 744 ) −80 .47 8 (96. 941 ) 106.6 84 ∗ (43 .1 54) − 1,2 59.009∗∗∗ (303.186) 47 8.857∗∗∗ ( 145 .710) 49 .0 94 (70.261) −16.8 64 (12.319) −86.821 (102.811) 113.051∗∗ (43 .827) − 1,2 85.796∗∗∗ (303.611) 5 14. 745 ∗∗∗ ( 140 .212) Fixed Effects... 2 3 4 302.5 ( 6,7 55) 207.5 ( 6 ,4 81) 95.0 (5.70) 41 9.2 ( 1,6 13) 223.7 ( 1,1 87) 195.5 (4. 08) 47 4.9 (590) 177.5 (625) 297 .4 (5.15) 1 ,4 44 .9 ( 34) 332.0 (31) 1,1 12.9 (6.79) Diff (2-1) (T-test) Diff (3-1) (T-test) Diff (4- 1) (T-test) 116.7 (3.79) 16.3 (0.51) 172 .47 (2.81) 30.03 (1.86) 1,1 42 .5 (7. 54) 1 24. 5 (4. 77) underlying collateral Panel C splits the sample into four levels of seniority (1 = most senior, 4 =... Market-toBook Profitability Leverage −3 34. 580∗∗∗ (103. 743 ) −271.665∗∗ (110.191)