Merton Miller.




Advice to the (e-)reader



FAMe is available in pdf form, html form, and epub form. Each format has advantages and disadvantages:

In both browser html and ereader epub format, it is the client program and not the FAMe team that handles font-size, line-breaking, and page-breaking. This is often but not always good. For example, tables and figures may be broken at very inopportune spots. Fortunately, when the reader resizes the content on a strange-looking page, the page often suddenly looks great. (linux users: please install msttcorefonts or ttf-mscorefonts-installer.)

Academic articles often include tables that require a wide page for comprehension. For this reason, FAMe articles are not well-suited to reading on small-screen devices. A 10-inch diagonal high-resolution tablet screen is recommended, although a 7-8-inch high-resolution diagonal screen size may be acceptable. (Also, note that wide-screen is great if your e-reader does not decide to abuse it for two-column reformatting, in which case it becomes narrow-screen.) Please, do not think about reading FAMe on a 3.3-inch diagonal iphone. It would be an exercise in frustration.

Small text font-sizes often work better with wide tables, but large font-sizes often look better on the main text in the absence of tables and figures. If you can easily switch font-size on the fly, you can get the best of both worlds.

FAMe editorial board
Executive Editor
Bhagwan Chowdhry, UCLA Anderson School
David Aboody, UCLA Anderson School
Amit Goyal, Swiss Finance Institute, University of Lausanne
Ivo Welch, UCLA Anderson School
Associate Editors
Antonio Bernardo, UCLA Anderson School
Bruce Carlin, UCLA Anderson School
Kent Daniel, Graduate School of Business, Columbia University
Shaun Davies, Leeds School of Business, University of Colorado
Andrea Eisfeldt, UCLA Anderson School
Mark Garmaise, UCLA Anderson School
Campbell Harvey, Duke Fuqua
David Hirshleifer, UC Irvine Merage
Ravi Jagannathan, Kellogg School, Northwestern University
Hanno Lustig, Stanford University
Brett Trueman, UCLA Anderson School
Brian Waters, Leeds School of Business, University of Colorado
Production Editor
Lily Qiu


Submissions, advertising, and support



Advertisements and support

For advertisement inquiries, please email fame.jagazine@gmail.com.


FAMe's goal is to broaden the impact of academic finance and accounting research. Creating an issue of FAMe costs about $50,000, largely financed out of pocket by ourselves. We depend on the goodwill of authors, readers, and supporters. We cannot spread costs over a large subscriber base—we are not The Economist.

To help underwrite our efforts, please consider purchasing a $50 or $100 annual subscription, especially if you can cover this subscription expense from your research budget. There is a paypal button on our website (click here). The paypal receipt should make it easy to be reimbursed as a research expense. The library subscription fee is the same price as the individual subscription fee (and paid the same way). The $100 subscription fee entitles libraries to make the content available to their patrons online.


Instructions to authors

To submit a FAMe version of your article recently published or forthcoming in a “pre-approved” finance journal, please send your proposed short memo to: famesubmission@gmail.com. For papers from accounting journals, please use famesubmission+a@gmail.com, instead. For more information and detailed instructions on how to prepare a MeMO, please refer to FAMe guidelines for authors on the www.fame-jagazine.com website for more information. Read our most recent issue to understand our “flavor”—we are still evolving, too. Detailed instructions for accepted submissions are posted at http://www.fame-jagazine.info/.



If you have contacted us at one of the above email addresses (fame.jagazine@gmail.com), e.g., famesubmission@gmail.com, and have not heard back from us, please feel free to prod us again at ivo.welch@gmail.com and bhagwan@anderson.ucla.edu.

Featured MeMos


Bhagwan Chowdhry, Executive Editor
Taxes and corporate policies: Evidence from a quasi-natural experiment
Craig Doidge and Alexander Dyck
Common Advisers in Mergers and Acquisitions: Determinants and Consequences
Anup Agrawal, Tommy Cooper, Qin Lian and Qiming Wang
Do independent directors cause improvements in firm transparency?
Christopher S. Armstrong, John E. Core, and Wayne R. Guay
Money Left on the Table: An Analysis of Participation in Employee Stock Purchase Plans
Ilona Babenko and Rik Sen
Do Going Private Transactions Affect Plant Productivity and Investment
Sreedhar Bharath, Amy Dittmar, Jagadeesh Sivadasan
Do Acquisitions Relieve Target Firms' Financial Constraints?
Isil Erel, Yeejin Jang, and Michael S. Weisbach
Detecting News in Aggregate Accounting Earnings: Implications for Stock Market Valuation
Panos N. Patatoukas
Financial Flexibility, Risk Management, and Payout Choice
Alice Adams Bonaime, Kristine Watson Hankins, and Jarrad Harford
Distracted Directors: Does Board Busyness Hurt Shareholder Value?
Antonio Falato, Dalida Kadyrzhanova, and Ugur Lel
A Crisis of Banks as Liquidity Providers
Viral V. Acharya and Nada Mora
Labor Protection and Leverage
Elena Simintzi, Vikrant Vig, and Paolo Volpin
Household Debt and Social Interactions
Dimitris Georgarakos, Michael Haliassos and Giacomo Pasini
Duration of Executive Compensation
Radhakrishnan Gopalan, Todd Milbourn, Fenghua Song, and Anjan V. Thakor
Beauty is in the eye of the beholder: The effect of corporate tax avoidance on the cost of bank loans
Iftekhar Hasan, Chun-Keung (Stan) Hoi, Qiang Wu, and Hao Zhang
Broad-Based Employee Stock Ownership: Motives and Outcomes
E. Han Kim and Paige Ouimet
Syndicated Loan Spreads and the Composition of the Syndicate
Jongha Lim, Bernadette A. Minton, and Michael S. Weisbach
Elections, political competition and failure
Wai-Man Liu and Phong Ngo
The market value of corporate votes: theory and evidence from option prices
Avner Kalay, Oguzhan Karakas, and Shagun Pant
Outside Directors and Board Advising and Monitoring Performance
Kyonghee Kim, Elaine Mauldin, and Sukesh Patro
The Determinants of Recovery Rates in the US Corporate Bond Market
Rainer Jankowitsch, Florian Nagler and Marti G. Subrahmanyam
Shadow banking and bank capital regulation
Guillaume Plantin
Does the Tail Wag the Dog?: The Effect of Credit Default Swaps on Credit Risk
Marti G. Subrahmanyam, Dragon Yongjun Tang, and Sarah Qian-Wang
Debtholder Responses to Shareholder Activism: Evidence from Hedge Fund Interventions
Jayanthi Sunder, Shyam V. Sunder, and Wan Wongsunwai
Board expertise: Do directors from related industries help bridge the information gap?
Nishant Dass, Omesh Kini, Vikram Nanda, Bunyamin Onal, and Jonathan Wang
Bhagwan Chowdhry, Executive Editor
The Mert Miller Issue

Issues 3 and 4 of FAMe are AFA Presidents issues. You will see pictures of past AFA presidents, both when they were young (with permission) and when they were adults. Click on the pictures to find out more about them.

The production of FAMe is a love of labor for us. We are looking for support, funding and sponsorship to be able to delegate some tasks to paid hired staff editors. Right now, we have to do everything ourselves. With more support, we could increase the frequency of publication for FAMe. Of course, this also depends on the desire of scholars to continue to write and submit memos.

We are dedicating Issue 4 of FAMe to the memory of Merton Miller.


Bhagwan Chowdhry

Executive Editor





in pdf, html, and ebook format!

Craig Doidge and Alexander Dyck
Taxes and corporate policies: Evidence from a quasi-natural experiment
Journal of Finance | Volume 70, Issue 1 (Feb 2015), 45-89

Nobel Laureate Eugene Fama recently stated that the big open challenge in corporate finance remains to produce evidence on how taxes affect market values and thus optimal financing decisions (ARFE, 2011). We document important interactions between tax incentives and corporate policies using a “quasi natural experiment” provided by a surprise announcement that imposed corporate taxes on a group of Canadian publicly traded firms. The announcement caused a dramatic decrease in value although prospective tax shields partially offset the losses, adding 4.6% to firm value. In response to changing tax incentives, firms subsequently adjusted corporate policies. They increased leverage to gain interest tax shields and reversed changes in other policies made to capitalize on tax benefits. Because we document changes in valuations and in several intertwined corporate policies, our evidence is supportive of the view that taxes are important for corporate finance.

Income trusts and the tax policy change

The tax policy change we exploit is officially known as the Tax Fairness Plan (TFP) but is unofficially referred to as the “Halloween Massacre” because it took place after markets closed on October 31, 2006. The TFP affected a large number of publicly traded Canadian firms called income trusts. Similar to Real Estate Investment Trusts (REITs), firms that adopted the income trust structure could avoid paying almost all corporate taxes while retaining the advantages of the corporate structure. The trust structure, which arguably had no nontax rationale, allowed trusts to pass all income through to investors who were then taxed at the personal level. In contrast, income earned by public corporations is taxed twice, once at the corporate level and again at the shareholder level when income is distributed. Thus, income generated by trusts faced fewer and lower taxes than income generated by corporations, with the tax gain from holding a trust depending on the investor's personal tax status.

The trust market grew rapidly in the early 2000s. Established publicly traded corporations from a broad cross-section of industries converted to the trust structure and new firms went public as trusts. At its peak just prior to October 31, 2006, the trust market included 216 trusts from a wide variety of industries worth $165 billion (Canadian dollars) and four corporations worth another $88 billion had announced plans to convert but had not completed the conversion. In total, these firms accounted for almost 13% of the market value of the Toronto Stock Exchange. The widely held view at the time was that the tax-advantaged status of income trusts was here to stay.

Not only was the plan to tax trusts a surprise to the market, the tax change was dramatic (the corporate tax rate increased from 0% to 31.5%), and was not contaminated by other information or policy changes. It effectively eliminated the tax advantage of the income trust structure (REITs were exempted). Existing trusts were given a four-year transition period and could retain their preferential tax status through January 1, 2011. Following the TFP, funds stopped flowing into the trust sector and most trusts made new organizational choices.

Taxes and the value of corporate policies

Figure 1 shows the cumulative abnormal returns on a value-weighted portfolio of trusts around the TFP announcement. It confirms that the TFP announcement was a complete surprise and shows the dramatic decrease in trust values after the announcement. On average trust values fell by 15% although there was substantial variation across trusts. We use this variation to provide new market-based evidence on the value of tax shields and the extent to which taxes reduce firm value.

1: Returns around the Tax Fairness Plan announcement
This figure shows cumulative abnormal returns on portfolios of trusts and REITs from September 26 to December 29, 2006 (days -25 to +40 around the October 31 announcement). It also shows cumulative returns on a value-weighted portfolio of corporates.

The trust structure provided a tax shield until it expired at the end of 2010. After that, trusts could use debt or nondebt tax shields to lower their taxes. We expect the value drop following the TFP should be smaller for trusts with access to more potential tax shields. To test this idea, we regress each trust's value drop on its prospective tax shields, measured as the sum of debt and nondebt tax shields used by corporations in the same industry. We also control for other factors that could influence the value drop, including the value of the four-year transition period for each trust. In these regressions, we focus on the 149 trusts with complete data. As expected, trusts with more prospective tax shields were less affected by the TFP. Our estimates imply that a trust that is fully affected by the TFP and has mean prospective tax shields is worth 4.6% more compared to a trust with no prospective tax shields.

The tax gain from holding a trust was greatest for tax exempt investors (e.g., investments held in individual investors' retirement accounts and pension funds) and smallest for taxable Canadian investors. Therefore, trusts that have a marginal investor with a lower personal tax rate should be more affected by the TFP and have a greater value drop. We include a well-known proxy for the tax status of the marginal investor (the Elton-Gruber (1970) ex-dividend day drop ratio) in our regressions and find that this is the case. With this proxy, we can also test whether the value of prospective tax shields differs depending on the tax status of the marginal investor. We find that prospective tax shields are worth more if the marginal investor has a lower personal tax rate.

Finally, because the TFP was a change in the corporate tax rate (from 0% to 31.5%), we can directly measure the net impact of this change on firm value. Our estimates, which control for a variety of factors including the transition period and tax shields, imply that firm value fell by 17.5%.

Taxes and changes in corporate policies

Our evidence shows that taxes significantly reduce value. Thus, firms should be willing to change their corporate policies or organizational structure to seek advantageous tax treatment. If the benefits from advantageous tax treatment are large enough, firms should be willing to incur costs on other margins to gain preferential tax treatment, as long as there is a net benefit. This view suggests an interconnected nature across a range of corporate policies. The same predictions go in reverse if the preferential tax treatment is eliminated.

If the tax advantage of trust status provided incentives to alter corporate policies, the effects should be observable before the TFP as firms converted from corporate to trust status and after the TFP when trusts lost their tax advantage. We start with leverage, an important corporate policy with clear tax incentives. Prior to the TFP, debt had no value as a tax shield for trusts. As expected, we find that trusts had lower leverage than when they were organized as corporates. With the loss of the trust tax shield due to the TFP, trusts could increase leverage to create an alternate source of tax shields. To identify a change in leverage in response to the TFP, we estimate regressions that compare year-to-year changes in leverage for trusts relative to corporates from 2007 to 2010. After controlling for observable and unobservable firm characteristics, plus industry and year effects, we find that trusts increased their leverage by six percentage points relative to corporates, an economically significant change given that the average debt-to-value ratio was about 20% in 2006.

We also examine other policies more closely tied to the specific trust tax rules. To fully capture the tax benefits of the trust structure, trusts had to pay out all earnings. This made it costly to build up cash holdings and to use these holdings to finance investment. Therefore, trusts had a tax incentive to increase payout and decrease cash holdings, and to the extent that external finance is costly, they had a disincentive to invest. The evidence is consistent with these predictions. Prior to the TFP, trusts had higher payout, lower cash holdings, and lower investment than when they were organized as corporates. After the TFP, they reversed these changes by reducing payout, increasing cash holdings, and increasing investment relative to corporates. Although the changes in these policies depend on specific income trust tax rules, our analysis suggests that managers are willing to make substantial, and potentially costly, accommodations to a number of corporate policies to access preferential tax treatment. Most interesting are the investment results as investment policy is central to value creation.

As a final test of the interactions between tax incentives and corporate policy choices, we examine acquisitions. Prior to the TFP, trusts' tax-advantaged status increased their value relative to corporates. We predict that trusts were more likely to be acquirers and less likely to be targets compared to corporates. Similarly, with the loss of the tax advantage due to the TFP, we predict that trusts were more likely to be acquired after the TFP as takeovers are one channel that can change organizational form and potentially offer more tax shields. The results are consistent with these predictions.


We offer new market-based evidence on the value of tax shields as well as new evidence of interactions between tax incentives and corporate policy choices. The combination of event study and time-series evidence that we provide is difficult to reconcile with nontax explanations. The increase in leverage after the TFP is consistent with the importance of debt tax shields while the results for investment and acquisitions highlight the potential for tax incentives to alter value-relevant policy choices. To the extent that trusts made potentially inefficient and costly changes to their investment policies, the net tax benefit of being a trust was likely worth less than the potential gross benefit of 31.5%. The idea that firms are willing to make costly changes to important policies to gain preferential tax treatment is consistent with all our other evidence that suggests taxes have an important impact on corporate decision-making.


Anup Agrawal, Tommy Cooper, Qin Lian and Qiming Wang
Common Advisers in Mergers and Acquisitions: Determinants and Consequences
Journal of Law and Economics | Volume 56, Issue 3 (Aug 2013): 691-740

We find that common M&A advisers are generally chosen in ways that make economic sense. After controlling for other variables and accounting for the endogenous choice to use common advisers, deals with common advisers take longer to complete and provide lower bid premiums. We also find some evidence of lower target valuations, lower bid premiums and higher bidder returns in such deals. The magnitudes of some of these effects are quite substantial. For example, in common adviser deals, deal valuation (measured by Deal value/EBITDA) is lower by about 6 compared to the sample mean of 34.12, and bid premiums are lower by about 21 percentage points.

While we find no significant difference between the two types of deals in measures of deal quality (combined announcement returns and post-acquisition performance), our evidence that common adviser arrangements turn out to be somewhat better for acquirers than for targets favors the conflict of interest hypothesis over the deal improvement hypothesis about the role of common advisers in mergers and acquisitions. Why then do targets agree to share advisers? Comparing the observed probability of common adviser deals to their predicted probability based on random pairing of advisers and clients, we find no evidence that merging firms avoided sharing advisers during the 1980s, but strong and growing evidence of such avoidance during the ensuing two decades.

When the New York Stock Exchange (NYSE) merged with Archipelago Holdings, Inc. in 2005, Goldman Sachs served as the lead mergers and acquisitions (M&A) adviser to both parties. While such a dual advisory role is fraught with potential conflict, it can also improve deal quality. What is the nature of the conflict faced by a common M&A adviser? A common adviser receives advisory fees, which are usually contingent on deal completion, from both sides. That gives it a strong incentive to complete deals, even if they are ill-conceived, and to complete them faster compared to separate advisers. And a common adviser has an incentive to favor an acquirer over the target because the acquirer is the larger, surviving firm that can give the adviser investment banking (IB) business in the future. So the conflict of interest effect implies that relative to deals with separate advisers, common adviser deals should have: (1) lower time to deal resolution, (2) lower deal quality, (3) lower target valuations, (4) lower acquisition premiums, and (5) lower (higher) wealth gains to target (acquirer) shareholders.

Two forces can keep a common adviser from responding to its incentives, potentially reducing the effects of the conflict. First, given the repeat nature of its business, the adviser may be deterred from exploiting its clients by the fear of damage to its reputation and potential litigation costs. Second, unlike buyers and sellers who sometimes share a realtor in the residential real estate market, merging companies are sophisticated entities run by managers, overseen by boards and advised by legal counsels, which likely consider the adviser's conflict before deciding to share an adviser with the counter-party and acting on its advice.

But a common M&A adviser can also use its dual role to improve deal outcomes. An M&A adviser, usually an IB, contractually agrees to 'aid and assist' a client through the M&A process. During this process, an unshared adviser working for the seller (buyer) receives information provided by the buyer (seller), but it is likely to be less than the information that the buyer (seller) makes available to its own adviser. Presumably, a common adviser has greater access to information compared to the information available to an unshared adviser. A common adviser also has greater control over the timing of information exchanges between a target and an acquirer. As information conduits, common advisers can improve deal outcomes by reducing information asymmetry between acquirers and targets. This deal improvement effect does not have a clear implication for the time to deal completion, nor does it imply that the adviser favors one party at the expense of the other. So this effect implies that, relative to deals with separate advisers, common adviser deals should have: (1) higher deal quality, (2) similar or higher target valuations, (3) similar or higher premiums to targets over their pre-bid share prices, and (4) higher gains from the merger to both target and acquirer shareholders.

The two effects of a common adviser, conflict of interest and deal improvement, are not mutually exclusive. Which effect dominates is an interesting empirical question that we address in this paper. We start by examining the determinants of the choice to use a common or separate M&A adviser. We then examine the consequences of this choice on several deal outcomes such as the speed of deal completion, deal quality, target valuation multiples, bid premiums, and the announcement returns to targets and acquirers. We analyze a sample of 6,272 acquisitions of U.S. targets by public acquirers during the period 1981-2005. Of these, 98 deals have common advisers and the remaining have separate advisers. Though rare, deals with common advisers are economically important, with an average deal size of $982 million and total deal value of $96 billion in inflation-adjusted 2005 dollars. And many of these deals involve prominent companies.

We account for the endogenous nature of the choice to use common or separate advisers by using four econometric approaches: Heckman's (1979) treatment effect model, two-stage least squares (2SLS), propensity score matching (PSM) and Abadie-Imbens (2006) matching (AIM). In addition to analyzing the full sample, we employ all of these methodologies on a choice-based sample designed to address the estimation issues that arise from common adviser deals being rare events.

Determinants of the choice of a common adviser

If the choice to use a common adviser is a random occurrence, we should not expect it to be related to firm, deal and adviser characteristics. Regression results show that this choice is generally made in economic sensible ways. Specifically, a common adviser is more likely to be picked in deals that are larger relative to the acquirer and smaller in absolute size; deals that involve private targets or stock payment; deals where each party has a prior relationship with the counter-party's, but not its own, adviser; deals where both parties are advised by a top adviser and multiple advisers; and deals where more IBs specialize in industries of both target and acquirer.

Are deals with common advisers resolved more quickly?

We estimate regressions of the natural logarithm of the number of days to deal completion, where the main explanatory variable is an indicator variable for a common adviser deal. The regressions control for other things. After accounting for selectivity, we find that the use of common advisers increases the time it takes to complete a deal. The magnitude of this increase ranges from 1.6 days in the treatment effect regression to 38.7 days under the PSM approach. Compared to the sample mean of 121.8 days to deal resolution, the marginal effect of having a common adviser represents an increase of 1.3% to 31.8%. This result does not support the conflict of interest effect that common advisers use their influence and information advantage to hurry up the M&A process.

Deal quality

Following prior research, we measure deal quality by the value a deal creates over the short-term and long-term for the shareholders of acquirers and targets. For acquisitions of public targets, we compute the combined cumulative abnormal return (CCAR) around acquisition announcement dates to measure the short-term value created by an acquisition. We measure the long-term value created by an acquisition as the post-acquisition stock performance of acquirers. Our regressions find no evidence that having a common adviser affects the short- and long-term value that a deal creates for the shareholders of the combined company.

Valuations, premiums and announcement returns

We measure target valuation using two multiples, computed as deal value divided by the target firm's sales or EBITDA, and acquisition premium as  100  times [( text{Deal Value} /  text{target's equity market cap at event day –40}) - 1] . Our regression results suggest that the use of common advisers leads, on average, to deals with lower valuations and lower acquisition premiums for targets. The magnitude of this effect is about 6 (or 18% of the mean valuation) for Deal value/EBITDA, and about 21 percentage points in lower acquisition premium. This evidence suggests a net conflict of interest effect: common advisers appear to favor acquirers at the expense of targets. Our regression results provide some evidence that deals with common advisers lead to higher acquirer CARs (by about 6 percentage points), while target CARs are unaffected.

Why do targets agree to share advisers?

We find that, on average, deals with common advisers work out better for acquirers (who experience higher abnormal returns upon deal announcement) than for targets (who receive lower deal valuations and bid premiums). These findings favor the conflict of interest hypothesis over the deal improvement hypothesis about the role of common M&A advisers. Why then do targets agree to share advisers with acquirers? Well, a priori, it is unclear whether common advisers will respond to their incentive to favor acquirers over targets in the face of advisers' reputational and litigation concerns, and the benefit from potential deal improvement may outweigh any cost to targets from adverse incentives of common advisers. To shed some light on this issue, we test whether merging firms avoid sharing advisers. We do this by comparing the actual probability of common adviser deals to the predicted probability of such deals assuming purely random choice of advisers by acquirers and targets. A finding that firms avoid (seek out) [neither avoid nor seek out] common advisers would suggest that the conflict of interest (deal improvement) [neither] hypothesis dominates.

Except for three years during our sample period, the predicted probability of common adviser deals consistently exceeds the observed probability. During 1981-89, the difference between the predicted and actual probabilities is statistically insignificant. But the predicted probability exceeds the actual probability by 2.4 percentage points during the 1990s and by 3.2 percentage points during 2000-05. Both differences are statistically significant. The results suggest that while merging firms did not avoid sharing advisers during the 1980s, they did so actively during the 1990s and 2000s.

Why? We can think of at least three possible explanations of this growing avoidance of common adviser deals. First, as targets realized that they usually end up getting the short end of the stick in common adviser deals, they start avoiding common advisers. Second, there is increasing sensitivity to (and scrutiny of) conflicts in the corporate sector and on Wall Street in the wake of numerous corporate scandals such as Enron, the adoption of Sarbanes-Oxley Act in 2002, and the global analyst settlement in 2004. Finally, the growing avoidance of common adviser deals may also be due to an increase in litigation costs over time. We find that deals with common advisers attract securities class action lawsuits with somewhat greater frequency than deals with separate advisers. We leave a complete resolution of this issue to future research.


We find that common M&A advisers are generally chosen in ways that make economic sense. After controlling for other variables and accounting for the endogenous choice to use common advisers, deals with common advisers take longer to complete and provide lower bid premiums. We also find some evidence of lower target valuations, lower bid premiums and higher bidder returns in such deals. The magnitudes of some of these effects are quite substantial. For example, in common adviser deals, deal valuation (measured by Deal value/EBITDA) is lower by about 6 compared to the sample mean of 34.12, and bid premiums are lower by about 21 percentage points.

While we find no significant difference between the two types of deals in measures of deal quality (combined announcement returns and post-acquisition performance), our evidence that common adviser arrangements turn out to be somewhat better for acquirers than for targets favors the conflict of interest hypothesis over the deal improvement hypothesis about the role of common advisers in mergers and acquisitions. Why then do targets agree to share advisers? Comparing the observed probability of common adviser deals to their predicted probability based on random pairing of advisers and clients, we find no evidence that merging firms avoided sharing advisers during the 1980s, but strong and growing evidence of such avoidance during the ensuing two decades.

Christopher S. Armstrong, John E. Core, and Wayne R. Guay
Do independent directors cause improvements in firm transparency?
Journal of Financial Economics | Volume 113, Issue 3 (Sep 2014), 383-403

Independent directors, as outsiders to the firm, must acquire and process a substantial amount of firm-specific information to effectively perform their advising and monitoring duties. Independent directors are less effective when the corporate information environment is opaque, and when there are significant costs to acquire and process detailed information about their firm's operating, financing, and investing activities. Although prior research has made some progress in understanding the correlation between board independence and corporate transparency, this research does not test for a causal effect of board structure on transparency. Specifically, the results in this prior literature do not discriminate between two competing explanations for a positive correlation between transparency and board independence. One possibility is that corporate transparency is dictated mainly by exogenous firm characteristics, and that firms are able to appoint a higher proportion of independent directors when transparency happens to be greater. A second possibility is that firms can make choices that improve transparency, such as attracting analysts to the firm and committing to certain disclosures, which can make it easier for independent directors to become informed about the firm's activities.

NYSE and Nasdaq regs require majority independent directors

Our objective is to determine whether and how corporate transparency adapts to the informational demands of a particular board structure. Our research design relies on regulations that mandated substantial increases in the proportion of independent directors for some firms but not others. Specifically, we use regulations issued in 2003 by the NYSE and Nasdaq that required most listed corporations to have a majority (more than 50%) of independent directors on their boards. In general, firms were required to comply with these regulations by the earlier of: (1) the listed firm's first annual shareholder meeting after January 15, 2004; or (2) October 31, 2004. Some firms already had a majority of independent directors on their boards and therefore complied with these new regulations at the time they were issued. Other firms, with boards consisting of a majority of inside or non-independent directors, were required to add more independent directors, remove some inside directors, or some combination of the two.

In our sample, the firms that were not in compliance with the majority board independence rule (as of 2000) have a 45% increase in the mean proportion of independent directors (from 40% to 58% independent directors), whereas firms that were already in compliance experienced virtually no change in their proportion of independent directors during the same period.

1: Panel A: Board characteristics for compliant and non-compliant firms
Compliant Non-Compliant
Mean Median Mean Median
Number of Directors (2000) 9.15 9.00 8.03 8.00
% Independent Directors (2000) 72% 71% 40% 43%
Min. Required % Change Independent Directors (as of 2000) 0% 0% 22% 18%
Actual Change in % Independent Directors (from 2000 to 2004) 2% 2% 18% 17%
Our sample is 1,396 (453) firms that were compliant (non-compliant) with the 2003 NYSE and Nasdaq board independence regulations as of 2000. The first two rows reflect the firms' situation as of 2000. The last row shows the actual change in the percentage of independent directors by the end of 2004.

We use a model of board structure to identify the expected change in proportion of independent directors based on the minimum change that firms would have to make to their board structure, if any, to comply with the exchange requirements. We then use the predicted change in the proportion of independent directors (over a four-year period from roughly 2000-2004) to identify the effect of this exogenous change in board structure on changes in a variety of information-related variables.

Table 1 also provides descriptive statistics on how firms alter their initial board structure to comply with the exchange listing requirements. Because compliance is based on the ratio of independent directors to board size, a non-compliant firm can comply by decreasing board size (and removing inside or gray directors) or increasing board size. We find that 145 of the non-compliant firms reduced the size of their board. These firms went from an average of 9.89 directors to 7.88 directors—an average decrease of 2.01 directors. This average overall reduction in board size was the result of adding 0.98 independent directors, but removing 1.63 inside directors and 1.36 gray directors. We also find that 308 of the non-compliant firms increased the size of their board from 7.14 to 8.35 directors, on average. This average increase in board size was the result of adding 2.01 independent directors, and removing 0.28 inside and 0.52 gray directors. Overall, there is substantial cross-sectional variation in how firms adjust the size and composition of their boards to comply with the regulations.

How can public information assist independent directors?

We emphasize that our predictions and tests focus on public measures of corporate transparency. One might question whether independent directors could instead resolve their information disadvantage vis-a-vis management using private rather than public channels. For example, independent directors have access to internal budgeting information, reports, and informal communication with managers. It seems unlikely, however, that outside directors rely solely on information supplied by, and filtered through, managers (Adams-Hermalin-Weisbach (JEL 2010), Armstrong-Guay-Weber (JAE 2010)). Although managers will be forthcoming in sharing certain types of information with independent directors, they are not likely to share information that is detrimental to their own interests. Thus, managers are expected to be forthcoming only with information that is relatively unhelpful to independent directors with monitoring.

In light of this concern, independent directors seek financial reporting systems and public information channels that aid their monitoring activities. Bushman-Chen-Engel-Smith (JAE 2004) note that public disclosures can carry greater credibility than private communications, in part, because these disclosures are subject to SEC rules and enforcement, litigation, and oversight by auditors and other stakeholders. Information intermediaries such as analysts and the business press scrutinize public disclosures. Analysts can also uncover distortions in these disclosures (Miller (JAR 2006)). Other governing entities, such as blockholders and institutional investors, use public disclosures to monitor managers' strategic decisions. Further, because publicly-released erroneous information imposes costs on managers, public information channels enhance the credibility of private information that managers share with directors. For example, a public management forecast of earnings enhances the credibility of non-public budgets that underlie the forecast. We conclude from these arguments and prior literature that public measures of corporate transparency are likely to play an important role in governance.

Do firms alter their information environment to assist?

Using board structure data for a broad sample of 1,849 firms, we find that bid-ask spreads (a common proxy for information asymmetry), decreases following an exogenous increase in the proportion of independent directors. The typical 18 percentage point average increase in the proportion of independent directors following the 2003 rule translates to roughly a 6% decrease in information asymmetry. We also explore some of the potential channels through which transparency may be altered to satisfy the informational demands of the board. Specifically, we examine variables related to management forecast frequency and precision, accrual quality, analyst following and consensus, shareholder base, and auditor fees. Our evidence suggests that an increase in the proportion of independent directors results in increases in the frequency of management forecasts and analyst following (with marginally significant evidence of increases in management forecast precision, analyst forecast consensus, and institutional following). We also find that auditor fees, a proxy for commitment to stringent internal controls and higher quality financial reporting, are somewhat larger at firms that are required to increase the independence of both their full board and their audit committee.

Are entrenched managers less likely to improve transparency?

We also explore several predictions regarding cross-sectional variation in the relation between increases in board independence and corporate transparency. For example, although a board with a majority of independent directors likely requires a more transparent information environment than an insider-dominated board to govern effectively, management may not willingly relinquish control over the board. One way for management to limit the monitoring efficacy of independent directors—particularly those who are new—is to withhold information or otherwise resist efforts by directors to elicit increased transparency. To explore this possibility, we test whether the increase in corporate transparency is muted for firms where managers are likely to be entrenched. Using insider ownership and the proportion of independent directors appointed during the current CEO's tenure as proxies for entrenchment, we find that when management is likely to be entrenched, management forecasts become less precise, and to some extent less frequent, following the mandated increase in board independence. However, we find no significant mediating effects of management entrenchment on the relations between the change in board independence and changes in our other measures of corporate transparency.

Our other cross-sectional analyses examine whether the relation between increases in board independence and corporate transparency varies with audit committee independence, information processing costs, and whether firms operationalize their compliance with the board structure regulations by adding new independent directors (thereby increasing board size) as opposed to removing non-independent directors (thereby decreasing board size). Our findings indicate that information asymmetry declines more for firms that increase their board size (by adding new independent directors), as well as for firms that are required to simultaneously increase the independence of their audit committee.

Do changes in transparency precede or follow board changes?

Finally, we explore the lead/lag relation between changes in transparency and changes in board independence. We partition our sample of non-compliant firms into those that complied early and those that complied late in the sample period, and find some evidence consistent with improvements in transparency both preceding and following compliance with the board independence regulations.


Collectively, our results suggest that firms can and do alter certain aspects of corporate transparency to facilitate the informational demands of independent directors. These results generally support the inferences in a large body of financial reporting and disclosure literature arguing that corporate transparency is endogenous with respect to management and/or board actions. At the same time, our findings also highlight the importance of acknowledging the simultaneous relation between board independence and corporate transparency, and we suggest that caution be exercised when interpreting results that take board independence or corporate transparency as exogenous rather than both being jointly and simultaneously determined.


Ilona Babenko and Rik Sen
Money Left on the Table: An Analysis of Participation in Employee Stock Purchase Plans
Review of Financial Studies | Volume 27, Issue 12 (Dec 2014), 3658-3698

Are individuals good at making investment choices? In academia, this question has been debated in a number of different contexts, such as participation in the equity market and saving for retirement.

However, because a wide variety of unobserved factors (e.g., individual risk aversion) can determine the optimal choices of individuals in these contexts, it is often difficult to find definitive evidence of suboptimal behavior. Additionally, the interpretation of evidence often depends on the specification of the presumed rational behavior, on which opinions may vary.

We avoid these issues by analyzing a unique setting provided by employee stock purchase plans (ESPPs) and empirically analyze a fundamental prediction of economic theory - that individuals should always take up an investment opportunity with substantial positive profits and zero risk.

What are employee stock purchase plans?

In essence, ESPPs are company-run programs that allow participating employees to buy company stock at a discount. In most plans, employees contribute through payroll deductions over a purchase period of several months. On the purchase date, the company uses the accumulated funds to buy shares at a discounted price. The discount is typically set at the 15% of the prevailing market price. Most ESPPs, however, have a feature called lookback, which sets the pre-discount purchase price to the lower of the market price at the time of the purchase and the price at the beginning of the purchase period. The lookback increases substantially the potential benefit of ESPP participation. A unique interesting feature of ESPP plans is that they often allow participating employees to sell the stock immediately after purchasing it, and our analysis focuses only on such plans. This gives employees an opportunity to secure a substantial profit without taking any downside risk - they can buy stock at a discount from the market price and sell it immediately at the market price.

ESPP participation rates are low and employees leave a lot of money on the table

Most employees fail to take advantage of the money-making opportunity provided by ESPPs. In our sample of large publicly traded U.S. firms (S&P 500, Midcap 400, and Nasdaq 100), the average participation rate is below 31%. The distribution of participation rates in Figure 1 reveals that they are lower than 80% for almost all firms, and are less than 10% for quite a few firms.

1: Distribution of ESPP participation rates across firm-years
ESPP participation rates are less than 80% for almost all firms and less than 10% for quite a few firms.

Employees who do not participate in the plan leave a considerable amount of money on the table, forgoing, on average, $3,446 each year. Even after we account for transaction costs, taxes, and the possibility of employee separation from the firm, the average employee loss is still equivalent to an annual salary increase of $3,079, which is approximately 2.8 weeks of pay. These numbers translate to aggregate employee losses of over $7 billion per year across the 239 firms in our sample. If we make an additional extreme assumption that all employees are liquidity constrained and would need to borrow on their credit cards at a 14% interest rate in order to participate in an ESPP, we still find that an average non-participating employee forfeits a value of $2,877.

What factors are related to participation in ESPP?

We use data at two levels of aggregation to analyze employee participation decisions. The first data set has firm-level data on employee stock purchase plans obtained from SEC filings during 1998-2009. ESPP participation is defined as the contributions to the ESPP per employee normalized by the maximum allowed contribution. The second data set comes from a survey and has detailed information about individual employees, including their salaries, wealth, investments in the equity market, 401(k) plans, and whether an employee ever participated in a company ESPP.

We first analyze factors that should matter if employees strictly maximize their wealth, have the ability to understand multiple options, possess perfect self-control, and choose rationally. In particular, we consider plan attractiveness, the opportunity costs of time, and employee liquidity constraints. We then move on to factors that can be important if employees have limits to processing information. In this context, we consider awareness of the plan, familiarity with trading stocks, financial literacy, trust, loyalty, and past individual experiences in the stock market.

Overall, the participation rates tend to be higher in firms with more generous plans (Table 1). A 1% increase in the discount offered by the plan leads to an increase in participation of 0.8%. The corresponding effect for value of the lookback option is smaller at 0.3%. This can be explained by preference of employees for sure gains over risky gambles or by the failure to estimate the value of the lookback option. ESPP participation also increases with time passage since the plan adoption, suggesting that employee awareness of the plan is important for participation. Employees of financial firms are likely to have a higher level of financial literacy and should understand better the benefits of participating in an ESPP. Participation rate is indeed 8.2% higher for employees in financial firms. We also find an effect of employee loyalty. Firms that are rated as one of the top 100 companies by the Great Place to Work Institute see a 12.9% higher participation in their ESPPs.

1: Employee Participation in an ESPP (Firm Level Data)
Survey salary 0.22 0.18
Years since adoption 0.58 0.63
Finance industry 8.20
100 best company 12.9
Discount value 79.6 75.9
Lookback value 34.3 36.5
Observations 2,059 2,059
The dependent variable is ESPP participation at the firm level.

Using the individual-level data, we find that participation is higher among people with higher salaries (Table 2). This can be explained by higher dollar benefits from participation (the contribution limit goes up with salary) or by lower liquidity constraints. The evidence also points to the importance of familiarity in dealing with stocks. For example, people who have experience trading other securities and those who report having ever received stock options are more likely to participate in an ESPP. These results can be interpreted as evidence that there are fixed costs associated with ESPP participation, such as figuring out plan details and tax treatment and opening a brokerage account, that can affect the decision to participate in an ESPP. We expect these learning-related costs to be higher for individuals with a lower level of education. Indeed, we find that education attenuates the relation between familiarity in dealing with stocks and ESPP participation.

2: ESPP Participation and Employee Characteristics (Individual Level Data)
BA degree 0.27 0.38
Log(salary) 0.23 0.22
Ever received stock options 0.82 0.82
Trade other securities 0.46 0.30
Positive stock return experience 0.27 0.45
BA degree*Trade other securities –0.21
BA degree*Positive stock return experience –0.27
Observations 7,453 7,453
The dependent variable is equal to 1 if an employee ever participated in the firm's ESPP. The model is estimated by probit with firm fixed effects and country fixed effects.

The results also underscore the importance of education and financial literacy for making participation decisions. For example, individuals with a bachelor's degree are 4% more likely to participate in the plan. Similarly, people who make mistakes in the valuation of their stock options are more likely to miss out on ESPP benefits. Participation is also related to the experience of the individual with the company stock. A one standard deviation increase in annual stock return over the course of an employee's tenure increases the probability of participation in an ESPP by 4.8%. For a rational individual, past returns should not matter for participation, because they are not reliable predictors of future returns and because the stock obtained through an ESPP can be sold immediately. We expect that more educated individuals are more likely to understand this irrelevance. In our RFS paper, we do find that the effect of past stock returns is significantly stronger for individuals without a bachelor's degree.


The majority of employees in large public U.S. firms do not take advantage of the riskless and profitable investment opportunities provided by employee stock purchase plans. Individual losses associated with this behavior are non-trivial, with the average employee forfeiting $3,079 per year by not signing up for the plan.

Non-participation is at least partially attributable to employee financial illiteracy and lack of familiarity in dealing with stocks. Employees who fail to take advantage of riskless investment through an ESPP are also less likely to enroll in a 401(k) plan or to participate in the broader equity market. Although ESPPs allow employees to sell the company stock immediately after purchasing it, over 45% of individuals never sell the stock over their tenure and maintain highly undiversified portfolios. These findings suggest that individuals are making mistakes that substantially affects their welfare.

Sreedhar Bharath, Amy Dittmar, Jagadeesh Sivadasan
Do Going Private Transactions Affect Plant Productivity and Investment
Review of Financial Studies | Volume 27, Issue 7 (July 2015), 1929-1976

Changes in Productivity of Private firms

Are private firms more efficient than public firms? Jensen (1986) suggests that going-private could result in efficiency gains by aligning managers' incentives with shareholders and providing better monitoring. In this paper, we examine a broad dataset of going-private transactions, including those taken private by private equity, management and private operating firms between 1981 and 2005. We link data on going-private transactions to rich plant-level US Census microdata to examine how goingprivate affects plant-level productivity, investment, and exit (sale and closure). While we find within plant increases in measures of productivity after going-private, there is little evidence of efficiency gains relative to a control sample composed of firms from within the same industry, and of similar age and size (employment) as the going-private firms. Further, our productivity results hold excluding all plants that underwent a change in ownership after going-private, alleviating the potential concern that control plants may undergo improvements through ownership changes.

Controlling for Decision to Go Private

A key issue is the endogeneity of the going-private decision, i.e., whether firms that choose to go private are different in some way, and whether these differences account for any observed changes after going private. If the factors driving the going-private decision are related to industry-size-age factors (e.g., if say larger firms in some industries were more prone to going private), the matching in the baseline analysis controls for this flexibly. To further address potential endogeneity concerns, we create additional matched samples using past plant productivity and the propensity to go private. Both the propensity and past productivity matched analysis continue to show that plants of firms that go private do not improve their productivity relative to their matched controls. Thus, our findings suggest that operational efficiency of establishments that went private are not differentially enhanced even six years after going-private. While public, these same establishments have either higher productivity or no difference relative to control establishments. These results serve to cast doubts on the popular view that going-private produces productivity gains in plant level operations by alleviating agency issues or overinvestment problems (Jensen (1986)).

Changes in Investment and Plant Exits

Next, we examine investment decisions and establishment exits. Specifically, we examine establishment-level capital stock, employment and plant exits through sales and closures. We find that firms shrink capital stock and employment in the six years after going-private. Specifically, going-private firms decrease capital by 15.1% and employment by 3.3% relative to a size, industry and age control group. Despite these decreases in inputs, there are no productivity gains since output also falls commensurately for these firms. We further find that going-private firms exit plants more quickly (15.3% higher hazard rate) than this same matched control group, particularly in the three years after going-private. The higher exit is mainly driven by greater sales (33% higher hazard rate) rather than by closures (6.1% higher hazard rate).

We finally test whether going-private firms target lower performing plants for sales and closures after delisting. We find that going-private firms are more likely to exit by selling plants that have lower productivity. Thus, the going-private firms appear to take actions that improve the productivity of their portfolio of plants, but this is achieved through selling low productivity plants rather than through productivity improvements of individual plants. Taken together, these results suggest that: (i) agency problems and other constraints such as short-termism associated with public capital markets do not affect operational plant efficiency; (ii) investors in going-private transactions potentially gain value not by improving productivity within target plants, but by identifying productive plants, selling and closing (less productive) plants, and reducing capital.

Results by Type of Going Private Transaction

To explore potential alternative explanations of our results, we examine differences by types of acquirers involved in going-private transactions—private equity firms, management, and private operating firms. We compare changes in these groups relative to the matched control group used in earlier analysis. One explanation for the findings in this paper is that the high leverage used or the time horizon for recouping investment in going-private transactions could lead to short-termism and suboptimal decision making after delisting, as well as pressure to downsize plants. Because higher leverage and the short investment horizon are more likely to impact management buy-outs and private equity takeovers, examining the outcomes for operating acquisitions allows us to consider a sub-sample where high leverage is less of a concern.

In most specifications, regardless of the mechanism used to go private, there is no improvement in productivity relative to the matched control group. Thus, operating firm takeovers, which constitute about 45% of our sample and will have less impact from change in leverage, do not improve efficiency after going-private. We also find that operating firms have 8.3% to 11.6% declines in capital in the six years after going-private but smaller and not statistically significant declines in employment. Further, operating firms show a significantly higher propensity to both sell and close establishments relative to matched controls; in fact, operating acquirers are more likely to close a plant relative to matched controls than either private equity or management acquirers in the six years after going-private. The propensity to sell is the highest for private equity acquirers relative to the control group, with a hazard rate over twice that of operating acquirers and six-times that of management acquirers. These qualitatively similar results for the operating firm sample and the overall sample suggest that leverage or investor short-termism is unlikely to be the main explanation for the baseline results.


Our paper contributes to the long literature that examines productivity changes around corporate events as well as the literature on going private. In addition to providing evidence that plant-level productivity is not adversely affected by the overinvestment problem due to agency conflict in listed firms (Jensen (1986) and Jensen and Meckling (1976)), this paper also sheds light on the potential for capital market myopia, as described in Stein (1989) and supported by evidence in Graham, Harvey and Rajagopal (2005) and Bhojraj et al (2009). If market myopia leads to underinvestment, we would expect a relative increase in capital stock, employment and a greater patience to exit under-performing plants after going-private. The myopia hypothesis would also predict that there would be productivity gains relative to public firms once the constraint is removed by going-private. We find no evidence that this is the case; thus, our evidence is largely inconsistent with public firms being more subject to myopia.


Isil Erel, Yeejin Jang, and Michael S. Weisbach
Do Acquisitions Relieve Target Firms' Financial Constraints?
Journal of Finance | Volume 70, Issue 1 (Feb 2015), 289-328

Firms sometimes face financial constraints and have to forgo valuable investment opportunities. Managers often claim that an important source of value in acquisitions is the acquiring firm's ability to relieve these constraints and help finance investment for the target firm. By being a part of a larger organization subsequent to an acquisition, the target firm can gain better direct access to capital markets through a parent firm, or can finance its investment from internally generated cash flows in other divisions. Consequently, the target will be able to undertake an increased number of profitable projects. Under such circumstances, acquisitions can create value.

In this paper, we evaluate the extent to which acquisitions lower financial constraints. This financing view of acquisitions predicts that prior to the acquisition, targets should be financially constrained, and that following the acquisition, the constraints should decline.

1: Target's Total Assets before the Acquistion (USD Million)
Completion No of Domestic Independent Private Public
year Deals Mean Median Deals (%) Targets(%) Targets(%) Acquirer(%)
2001 228 77 7.2 62.95 72.8 95.2 46.0
2002 393 126 7.7 59.07 69.7 96.7 44.3
2003 429 57 7.5 61.20 62.5 97.2 29.4
2004 603 79 8.0 62.35 70.0 97.2 38.5
2005 768 95 8.6 63.53 68.0 97.4 36.7
2006 999 49 7.2 60.20 68.8 97.7 35.3
2007 1270 38 7.3 65.51 77.5 97.7 35.0
2008 497 83 7.4 67.58 78.5 97.6 31.6
Total 5187 68 7.6 63.1 71.6 97.4 36.1

Data and Sample

Examining these predictions empirically is difficult, since for most acquisitions, one cannot observe financial data on target firms after being acquired. In addition, financial data on targets that are private and/or subsidiaries of other corporations are not publicly available in the U.S. In contrast to the U.S., most European countries require firms to report financial data publicly on an unconsolidated basis. Because of this disclosure requirement, we are able to construct a sample of 5,187 European acquisitions occurring from 2001 to 2008, each of which became a wholly owned subsidiary subsequent to the acquisition.

As documented in Table 1, most of the targets in our sample are private firms (97.4%) and they are quite small, with a median book value of assets of approximately $7.6 million. The descriptive statistics highlight that our study focuses on the firms for which an acquisition is a relatively attractive source of relieving financial constraints.

Measuring Financial Constraints in Target Firms

To assess whether financial constraints are alleviated by acquisitions requires that one can measure constraints in a particular target firm both before and after being acquired. While there are many ways to measure financial constraints, particularly useful ones come from observing managers' own actions regarding their financial position. When access to capital markets is imperfect, value maximization will lead managers to adopt financial policies that ensure that the most important investments continue to be financed. In this paper, we use the following three measures that have been suggested by the prior literature:

  1. Cash Holdings: Managers of firms that face difficulties in raising necessary capital will typically hold more cash as a precaution against coming up short in the future.
  2. Cash Flow Sensitivity of Cash: Firms should hold more of their incremental cash flow as cash if they fear that they may not be able to raises funds easily in the future.
  3. Cash Flow Sensitivity of Investment: A financially unconstrained firm should be able to undertake all valuable investments, while the quantity of investments a financially constrained firm will undertake will be an increasing function of its cash holdings.
Therefore, a decline in the target's cash holdings, its investment-cash flow sensitivity, and its cash-cash flow sensitivity following an acquisition would suggest that the target's financial constraints are reduced when a target is acquired. In addition, if the improvement in financing from acquisitions reduces financial constraints, investment should rise following the acquisition.

Main Results

Our empirical results suggest that acquisitions do mitigate financial constraints. Table 2 presents our main results.

We find that cash holdings, normalized by assets, decline by approximately 1.5% for an average target firm after being acquired. The sensitivity of cash to cash flow declines significantly from 10.4% to close to zero, which implies that the target firm goes from being constrained to unconstrained. Finally, there is a statistically significant decline in the sensitivity of investment to cash flow, with the magnitude of the post-acquisition sensitivity being less than half of that before the acquisition. All of these results are consistent with the view that acquisitions mitigate financial constraints, potentially providing a source of value by enabling target firms to improve their investment policy.

2: The effect of acquisitions on cash holdings, cash-cash flow and cash-investment sensitivities, and investment of target firms
Dependent Variable Cash/Asset δ(Cash/Asset) Investment/Asset Investment/Asset
After-Merger Dummy –0.0145** –0.0104** 0.0247** 0.0201**
(0.005) (0.004) (0.006) (0.005)
Cash Flow/Total Assets 0.1036** 0.0713**
(0.025) (0.028)
After-Merger Dummy × –0.0929** –0.0563***
Cash Flow/Total Assets (0.029) (0.033)
Firm-level and macro-level controls are always included
Observations 11,941 11,632 12,138 27,322
R2 0.672 0.195 0.368 0.315
Standard errors are in parentheses

We also find that investment does increase for target firms in our sample. Controlling for other factors, investment as a fraction of total assets increases by 2% following the acquisition, which is a substantial effect given that the mean (median) investment ratio is 6.4% (3.4%) for targets before the acquisition.

smaller targets. In addition, the reduction in financial constraints occurs in both diversifying and same-industry acquisitions. This cross-deal pattern of empirical results suggests that they reflect reductions in financial constraints rather than other factors.

Concluding Remarks

In this paper, we document that the financial management decisions of target firms change when the firm is acquired in ways consistent with their becoming less financially constrained. We find evidence that financial synergies resulting from reductions in financial constraints could motivate some acquisitions, potentially improving efficiency in investment policy of target firms.

While a reduction in financial constraints could potentially be a factor leading to acquisitions, it is not likely to be the only factor. Quantifying the relative importance of financing motivations compared to other factors is an important question for future research.

Panos N. Patatoukas
Detecting News in Aggregate Accounting Earnings: Implications for Stock Market Valuation
Review of Accounting Studies | Volume 19, Issue 1 (Mar 2014), 134-160

How much new information is there in accounting earnings? A long line of research dating back to Ball and Brown (1968) investigates the informativeness of earnings at the firm level using simple regressions of stock returns on earnings changes. The estimated slope coefficient is commonly referred to as the earnings response coefficient (ERC) and indicates how much response there is in stock prices for a change in earnings. Studies at the firm level typically employ the earnings change as a measure of cash flow news and, consistent with the fundamental equality that prices are discounted expected future cash flows, find evidence of a significantly positive ERC. In contrast to firm-level findings, recent research extending the investigation at the aggregate level provides puzzling evidence of a weak, or even negative, association between aggregate earnings changes and stock market returns. This result is puzzling because one would expect that positive aggregate earnings changes translate into favorable cash flow news and, in turn, higher stock market prices.

Are earnings informative?

The decomposition of Campbell (1991) provides the basis for elucidating the relation between earnings changes and stock returns. Campbell (1991) decomposes stock returns at time t into expected returns at time t-1 and revisions in expectations at time t about future cash flows (cash flow news) and future stock returns (discount rate news). With this decomposition in mind, it becomes clear that the relation between earnings changes and stock returns depends not only on the covariation of earnings changes with cash flow news but also on the covariation of earnings changes with the remaining components of realized stock returns.

In terms of this standard decomposition, literature in accounting and finance proposes two divergent views of the weak earnings-returns relation at the aggregate level. The first view, initially proposed by Kothari et al. (2006), suggests that aggregate earnings changes are informational. Specifically, aggregate earnings changes are mostly unanticipated and correlated with contemporaneous value-relevant news that causes investors to revise their expectations about not only future cash flows but also discount rates. A link between aggregate earnings changes and discount rate news implies that stock market returns can be weakly, or even negatively, related to aggregate earnings changes unless expected future cash flows increase by enough to offset an increase in expected future stock market returns.

In contrast, Sadka and Sadka (2009) propose that aggregate earnings changes are mostly anticipated and therefore provide little or no value-relevant news. This view suggests that aggregate earnings changes are non-informational and merely confirm investors' expectations. To explain a negative earnings-returns relation at the stock market level, they further argue that aggregate earnings changes are negatively related to lagged expected stock market returns. The argument goes as follows. Investors can anticipate aggregate earnings changes more than one year in advance, and as they predict higher future earnings, they demand a lower discount rate.

Yes, they are!

Which of the two views prevails is an open empirical question with important implications for the informational role of accounting earnings at the aggregate stock market level. My study directly addresses this question. I compile a comprehensive sample of U.S. publicly traded firms over the period from 1981 to 2009 and investigate the association of aggregate earnings changes with proxies for value-relevant revisions in expectations. I proxy for the unobservable ex ante expected future stock market return using the implied cost of capital method of Easton et al. (2002). The changes in the implied cost of capital estimates proxy for revisions in expectations about future stock market returns. I employ aggregate revisions in security analysts' projections of future rates of return on equity to proxy for revisions in expectations about future cash flows.

My study provides direct evidence that aggregate accounting earnings are correlated with new information that is relevant for valuation at the stock market level. First, I find a significantly positive association between aggregate earnings changes and revisions in expectations about future stock market returns. The link between aggregate accounting earnings and discount rate news extends across all components of the expected future stock market return. Specifically, revisions in expectations about the real riskless rate, future inflation, and the implied equity risk premium (i.e., the excess of the implied cost of capital over the nominal riskless rate) jointly explain as much as 28 percent of the time-series variation in aggregate earnings changes. I also find a significantly positive association between aggregate earnings changes and aggregate revisions in security analysts' projections of subsequent profitability rates. Aggregate analysts' revisions explain 48 percent of the time-series variation in aggregate earnings changes. In contrast, most of the time-series variation in aggregate earnings changes is not explained by lagged realized and expected stock market returns.

The link between aggregate earnings changes and new information about all components of the expected future stock market return implies that, univariately, the earnings-returns association can be weak, or even negative, at the aggregate level. This is because stock prices react to all value-relevant news (i.e., both cash flow news and discount rate news) embedded in accounting earnings. Indeed, a two-stage analysis reveals that over the sample period studied, cash flow news and discount rate news in aggregate earnings changes covary positively and have almost offsetting impacts on stock market prices. As a result, at the aggregate level, the simple ERC estimate is indistinguishable from zero. After I expand the right-hand side of the traditional ERC regression model to include contemporaneous revisions in expectations about all components of the expected future stock market return (i.e., the real riskless rate, expected inflation, and the expected equity risk premium), the multiple ERC estimate soars from virtually zero to +3.08 with a t-statistic of 3.59 (significant at the 1 percent level). The multiple ERC regression model also improves explanatory power with the adjusted R2 rising from below zero percent to 43 percent.

Aggregate earnings are informative

To address potential concerns about the construct validity of Easton's et al. (2002) implied cost of capital as a proxy for the ex ante expected future stock market return, I implement an instrumental-variables two-stage procedure. In the first stage, I estimate a regression of implied cost of capital on the real riskless rate, expected inflation, and a composite index computed as the first principal component of a comprehensive array of instrumental variables for the unobservable ex ante expected equity risk premium. Using the fitted value from this first-stage regression, I construct “cleaned-up” proxies for the expected future stock market return and discount rate news. In the second stage, I find that the multiple ERC estimate strengthens at +3.28 with a t-statistic of 3.89 (significant at the 1 percent level). In addition, the coefficient estimates from the multiple ERC regression model converge to theoretically predicted values when I directly take into account the time-series properties of aggregate profitability rates and discount rate components.

I conclude that, although the simple earnings-returns relation is weak at the stock market level, it is not the case that aggregate earnings changes are non-informational. Instead, if aggregate earnings changes are introduced “properly” along with new information about discount rates, then the earnings-returns relation improves considerably. Stated otherwise, the traditional ERC regression model fails to detect the stock market's reaction to value-relevant news embedded in aggregate accounting earnings because it suffers from a correlated omitted variables problem. Discount rate news is (1) positively associated with contemporaneous aggregate earnings changes and (2) negatively associated with contemporaneous stock market returns. My results highlight that researchers interested in detecting whether and how new information in aggregate accounting earnings maps into stock market prices need to carefully consider the overlap between aggregate earnings changes and revisions in expectations about all components of the future stock market return. My study contributes to research in accounting and finance by providing direct evidence on the association between accounting earnings and new information that is relevant for valuation at the aggregate stock market level. An interesting direction for future research would be to decompose aggregate earnings changes to identify which components are more correlated with value-relevant new information (e.g., revenues or expenses, accruals or cash flows). The findings are for all firms, on average. Further research may find cross-sectional variation in the link between earnings changes and news that is relevant for valuation at the stock market level. Such research would further broaden our knowledge of the link between accounting earnings and the aggregate economy.

Alice Adams Bonaime, Kristine Watson Hankins, and Jarrad Harford
Financial Flexibility, Risk Management, and Payout Choice
Review of Financial Studies | Volume 27, Issue 4 (Apr 2014), 1074-1101

Two key components of financial flexibility are payout policy and risk management. A firm's form of payout affects its financial flexibility: Because repurchases provide managers with more discretion than dividends in terms of the amount and timing of distributions, a payout structure favoring repurchases, relative to dividends, increases financial flexibility. Risk management also increases financial flexibility because hedging reduces cash flow volatility and can prevent underinvestment and financial distress. Thus, all else equal, smaller dividend payments relative to repurchases and lower cash flow volatility both enable a firm to invest when opportunities arise and to service outstanding debt obligations.

We use detailed data on a large sample of firms over an extended period of time and find that a more flexible payout structure offers operational hedging benefits. Firms that favor repurchases over dividends hedge less and vice versa. Using separate samples of financial and nonfinancial firms, we document that payout and risk management are interrelated decisions.

Measuring payout flexibility and risk management

To investigate the relationship between payout and hedging, we initially focus on a sample of bank holding companies from 1995 to 2008. Unlike other publicly traded companies, bank holding companies are required to report the level of derivatives use and to separate trading from hedging activity. Therefore, for a large sample of firms, we can examine directly whether the amount of hedging affects payout choices within firms. We measure hedging as the dollar value spent on interest rate derivatives, the most common derivative used by bank holding companies. Further, bank holding companies report the impact of derivatives on income and their interest rate exposure, the dominant hedgeable exposure. Hence, we can control for the risk profile of the firm using the income volatility calculated without the impact of hedging and the interest rate exposure. This ensures that hedging changes are not driven by changing risk exposures. Lastly, we define “payout flexibility” as the ratio of repurchases to total payout—although our results are also robust to alternative definitions.

Are payout flexibility and risk management simply related to underlying firm characteristics?

We begin by showing a strong negative relationship between hedging and payout flexibility in the cross-section. However, there is the concern that the same types of firms that engage in active risk management programs also pay regular dividends, which would result in a cross-sectional relationship between hedging and dividends but not support our premise that payout flexibility and hedging are jointly determined hedging tools. To rule out this possibility, we examine how the same firm trades off hedging and payout flexibility over time. Table 1 presents results for our main sample of financial firms, where we identify a negative and significant relationship between the level of interest rate hedging in bank holding companies and their payout flexibility.

1: The Substitution of Payout Flexibility and Hedging
Financial firms Nonfinancial firms
Dependent variable: Payout flexibility Hedging Payout flexibility Hedging
Hedging –0.259** –3.639**
Payout flexibility –4.418** –0.215**
Number of observations 12,628 12,402 19,661 19,767
Control variables Yes Yes Yes Yes
Firm fixed effects Yes Yes No No
Financial and non-financial firms substitute hedging for payout flexibility (repurchase/total payout) and vice versa.

We further verify our results using additional robustness checks. We use the state level adoption of Prudent Investor legislation as a shock to the cost of dividends and large increases in firm size as a shock to the cost of hedging. Lastly, we estimate a Heckman selection model to address the choice to hedge or distribute earnings, including both firm fixed effects and instrumental variables. All of our empirical analysis documents that payout and hedging are jointly determined, consistent with payout flexibility being a risk management device.

These findings are robust to controlling for the effect of hedging on cash flow volatility, the level of total payout, other measures of financial flexibility (including capital levels and cash ratios), and the presence of a major blockholder and firm size, which are commonly cited as determinants of hedging and payout decisions (e.g., Grinstein-Michaely (JF 2005);

We confirm our analysis using a panel of nonfinancial firms. Although the nonfinancial data are far less detailed, firms begin reporting gains and losses due to derivatives in 2004 which allows us to test whether our general conclusions extend to all publicly traded firms. Our results support the hypothesis that non-financial firms also substitute financial hedging and payout flexibility.

Paralleling the empirical work presented for bank holding companies, we find that nonfinancial hedging firms have significantly less flexibility in their payout structure, consistent with hedging substituting for payout flexibility. Table 1 includes the results from our instrumental variable model using nonfinancial firms. This indicates that a broad range of firms use payout flexibility as an operational hedge.


In sum, both bank holding companies and nonfinancial firms recognize that payout policy and risk management both contribute to financial flexibility and are substitutes. Consequently, payout policy—and the broader issue of financial flexibility—can be fully understood only within the context of the firm's related hedging choices.


Antonio Falato, Dalida Kadyrzhanova, and Ugur Lel
Distracted Directors: Does Board Busyness Hurt Shareholder Value?
Journal of Financial Economics | Volume 113, Issue 3 (Sep 2014), 404-426

Is independent director busyness detrimental to board monitoring quality? A large number of publicly traded firms in the United States have recently limited the number of multiple directorships held by their board members. For example, a recent survey shows that 74 percent of S&P 500 firms impose restrictions on the number of corporate directorships held by their independent directors, up from 27 percent in 2006, and Institutional Shareholder Services recommends restrictions on the number of multiple directorships. While several studies find that busy directors are associated with lower firm valuations and less effective monitoring others either do not, or provide mixed evidence. Using a quasi-natural experiment, we find that there is a negative valuation effect of “attention shocks” (exogenous increases in the demand for outside directors' time) and that several firm decisions for which board monitoring is material are also adversely impacted.

Our experiment

We hand-collected information on the deaths of directors and CEOs over the 1988 to 2007 period. We were able to find 633 independent director deaths, which resulted in a sample of 2,551 director-interlocked firms, 1,084 of which are due to sudden deaths. Our experiment constructs two groups of director-interlocked firms: a group of firms whose independent directors' committee workload increased—which is our “treatment group,” and a group of firms whose independent directors' workload did not increase—our “control group.” We offer direct evidence validating the notion that shocks originating from the death of either the CEO or a colleague on the board lead to an increase in board committee workload for directors in the treatment group, which takes away time and resources from their board activities at interlocked firms. We compare the change in director-interlocked firm value within the treatment and control groups before and after the death event dates (first-difference) and then take the difference across the two groups (second-difference).

Costly distractions?

Relative to the control group, firms in the treatment group experience a significant negative stock market reaction to director attention shocks. As reported in Table 1, for interlocked firms in the treatment group the market reaction is substantial when the death event is sudden. By contrast, in our control group of observations that are subject to the same sudden shocks but whose interlocked relation does not involve serving in the same committee, there is no statistically significant market reaction. The difference in the market reaction between the treatment and control groups is statistically significant and robust to a battery of tests that address potential outliers and sample composition issues.

The first row in Table 1 shows a negative and statistically significant decline in value for firms whose directors are interlocked through the same committee—i.e., those in our treatment group. Given the average market capitalization is $4.47 billion in this subsample, attention shocks result in a decline of $35.31 million. On the other hand, the market reaction is statistically indistinguishable from zero for firms where the interlocked relation does not involve serving in the same committee. The difference in market reaction between the treatment and control groups is statistically significant. Similarly, CARs within a (–15, +15) day window show a sharp decline for the treatment group of observations, whereas CARs appear to fluctuate around zero for firms whose directors do not share an interlocked relation through the same committee. When we focus on attention shocks that are the most unexpected in nature (i.e., due to sudden deaths of independent directors), the market reaction becomes even more severe.

1: Cumulative Abnormal Returns for Director-Interlocked Firms: Were interlocking and deceased directors on the same committee?
YES NO Difference
(treatment) (control)
All Deaths –0.794** 0.131 –0.925**
(0.286) (0.185) (0.335)
N 843 1,708
Sudden Deaths –1.554** 0.190 –1.744**
(0.554) (0.354) (0.645)
N 360 724
Firms values are lower when directors from interlocking firms on the same committee die, with stronger effects for sudden deaths. (Standard errors are in parentheses.)

When do distracted directors matter?

Table 2 reports results by various measures of the degree of constraints on directors' time within the treatment group. Smaller boards may have less slack in terms of directors' assignments into committees, and therefore may be affected more severely by an attention shock. When we split the sample based on the top versus bottom quartiles of board size (Small Boards), we find an average market reaction for firms with smaller boards, which is again statistically significant. More frequent meetings may be suggestive of greater workload to effectively get the tasks completed. In cases where the director is not already busy, the market reaction is muted.

2: Cumulative Abnormal Returns for Director-Interlocked Firms
Treatment group only, by quartile of X
Top Bottom Difference
X - 1/Board Size –1.515** 0.344 –1.859**
(0.554) (0.663) (0.858)
X - # Board Meetings –2.152** –0.370 –1.782**
(0.814) (0.411) (0.868)
X - % of Busy Drct –2.327** –0.163 –2.165**
(0.937) (0.448) (0.923)
X - # Outside Directorships –1.367** –0.139 –1.288**
(0.458) (0.466) (0.662)
X - Busyness Factor –2.157** 0.191 –2.348**
(0.761) (0.515) (0.903)
N 210 N 210
Firms' stock prices decline only when an already busy director receives an attention shock. (Standard errors are in parentheses.)


Our evidence indicates that independent directors' workload matters for investors. Our results also provide direct evidence supporting the too-busy-to-mind-the-business view of studies such as Fich and Shivdasani (JF 2006). As such, our evidence that additional demands on directors' time have adverse consequences for board monitoring quality and firm value suggests that multiple directorships can be detrimental to shareholder value especially when the independent director and the board are already busy. Thus, the recommendations of Institutional Shareholder Services to limit the number of seats directors can hold in publicly traded firms may benefit shareholders by making boards more resilient to adverse shocks that increase demands on directors' time.


Viral V. Acharya and Nada Mora
A Crisis of Banks as Liquidity Providers
Journal of Finance | Volume 70, Issue 1 (Feb 2015), 1-43

Our paper investigates whether the 2007-09 crisis was a crisis of banks as liquidity providers, reducing credit and increasing the fragility of the financial system. A widely accepted notion is that banks have a natural advantage in providing liquidity to businesses through credit lines and other commitments established during normal times. Banks are thought to function well as liquidity providers by combining deposit taking and commitment lending activities. Both activities require banks to hold liquid assets to provide liquidity-on-demand (Kashyap-Rajan-Stein (JF 2002)). Banks conserve on liquid asset buffers in meeting liquidity demands, provided deposit withdrawals and commitment drawdowns are not too highly correlated. Supportive evidence comes from previous crises such as in 1998 following the LTCM hedge fund failure. Even when market stress led to significant drawdowns, banks met the increased credit demand because of flight-to-safety flows from depositors as shown by Gatev-Strahan (JF 2006).

1: Deposit rates: Failed minus nonfailed banks
Weak institutions offered higher deposit rates in the run-up to failure.

In 2007 to 2009, however, the banking system was itself at the center of the financial crisis. Questions were raised about the solvency of the banking system as a whole, which was exposed to toxic credit instruments as documented by Acharya-Schnabl-Suarez (JFE 2013) and others. As the solvency risk of a bank increases, it might seek to attract deposits by offering higher rates such as Washington Mutual during the crisis: “The fact that Washington Mutual is now owned by Chase is very positive, because they were a huge outlier on rates” (Ken Lewis, then Chairman and CEO of Bank of America, cited in American Banker October 9, 2008).

Figure 1 shows this more generally by drawing on weekly survey evidence of deposit rates offered by failed banks. The figure plots the difference from the rates of banks that did not fail, over a one-year period prior to failure. The x-axis is the time to failure, where failure also covers cases of near-fails based on stock return performance. Weak institutions offered substantially higher CD rates in the run-up to failure. For example, the average difference on their 12-month CD rates reached close to 60 basis points. This finding aligns with the literature on market discipline where depositors identify insolvent banks and demand higher returns.

When the liquidity provision mechanism of the banking system as a whole breaks down

But the main result of this research is that the onset of the crisis was a crisis of banks as liquidity providers in the aggregate; not just of the weakest banks. It makes sense ex ante for banks to combine deposit taking with commitments, but banks may experience higher liquidity demand coming from both depositors and firms. It is important to understand how banks adjust to such a shock.

2: Market stress and net flows into deposits at commercial banks
As the VIX increased, deposits at commercial banks shot up.

Core deposits increased by only $90 billion up until end-2008Q2. This increase fell short of the increase in core deposits in a comparable period before the crisis and also fell short in relation to the surge in deposits in previous crises such as LTCM. More importantly, deposit inflows fell short in relation to increased loan demand and drawdowns during this period. Core deposits eventually increased in the banking system as a whole by close to $800 billion by early 2009, but only starting in 2008Q3 when they grew by $272 billion in just one quarter. Lending growth outpaced core deposit funding growth from the ABCP “freeze” in August 2007 to just prior to Lehman's failure. The aggregate loan-to-deposit shortfall is shown in Figure 2 below based on quarterly snapshots of weekly commercial banking data. For example, the cumulative difference between the increase in lending and in deposits reached $239 billion by end 2008Q2 and widened to more than $300 billion in the weeks just prior to Lehman's failure.

We argue that the weak deposit funding position of banks and its sharp reversal following Lehman's failure is explained by investor perception of greater risk in bank deposits relative to instruments offering similar liquidity and payments services. Investors piled into securities with more explicit government backing than bank debt, especially as most deposits were over the deposit insurance limit at the start of the crisis. Investors preferred government-backed securities because of concerns about the banking sector's health and the lack of information about exposures to the subprime shock. The rise in aggregate risk also reduced banks' ability to diversify shocks across business and deposits (Acharya-Almeida-Campello (JF 2013)).

For example, money market funds specializing in government securities funds saw greater funding inflows than prime funds. The flow into government funds accelerated in the aftermath of Lehman's failure and Reserve Primary Fund's “breaking the buck” when there was a run on a range of prime funds. The flight out of prime funds of $400 billion in the two weeks following Lehman's failure went to government funds as well as to deposits in the banking system of close to $200 billion as seen in Figure 2. Concurrently, the TARP backstop helped reverse the anemic deposit position of the banking system. TARP increased insurance limits from $100,000 to $250,000 and fully insured noninterest-bearing accounts, among other measures. The inflow of deposit funding enabled by explicit government backing finally allowed the banking system as a whole to close its loan-to-deposit shortfall.

Liquidity demand risk and individual bank behavior in the crisis

To understand liquidity provision by individual banks, we test whether a bank at greater risk of credit line drawdowns offers higher rates if it does not gain enough deposits to match its funding needs. The coefficient β2 on the interaction term of crisis and liquidity demand risk in the equation below is the one of interest. Because our thesis is predicated on the reversal in the aggregate liquidity shock in September 2008 and the primary data are quarterly Call Reports, crisis is empirically represented by two dummy variables, crisis1 (2007Q3-2008Q2) and crisis2 (2008Q3-2009Q2). We also confirm the findings by examining CD rates from a weekly proprietary survey. The key measure of a bank's vulnerability to liquidity demand is the ratio of unused commitments to the sum of unused commitments and on-balance sheet loans. Deposit Ratei,t = β1 · liquidity demand riski,t – 1  + β2 · liquidity demand riski,t – 1 × Crisist +   + Bank FEi + Time FEt + Other Controlsi,t + ei,t

We also test whether banks took additional actions to meet increased drawdowns. Banks will likely be forced to adjust by cutting back on new credit if deposit funding and interbank borrowing are not sufficient. We test for both claims. Other backup actions include running down liquid assets and seeking government-sponsored borrowing such as advances from Federal Home Loan Banks (FHLBs).

Our results show that the aggregate liquidity shock particularly hit banks at greater risk of credit line drawdowns. For example, a 0.1 increase in a bank's unused commitment ratio (roughly a 1 standard deviation or the difference between the 75 and 25 percentiles) raised a bank's large-time deposit rate by 5.9–7.1 basis points in the first year of the crisis. But crucially, despite scrambling for deposits by raising rates, commitments- exposed banks experienced weaker deposit growth. Even the growth of deposits that are commonly considered a stable source of funding such as core deposits was limited.

That banks honored their existing credit lines drawn by firms beginning in August 2007 was possible only because of explicit, large support from the government and government-sponsored agencies such as FHLB advances and Federal Reserve liquidity facilities. For example, advances from the FHLBs covered 65% of non-deposit borrowing growth at commitments-exposed banks during the first year, and the widening shortfall between their on-balance sheet loans and deposits was closed halfway with government-sponsored borrowing.

We conduct several tests to rule out the alternative hypothesis that commitments-exposed banks were simply those with greater solvency problems. The results indicate that solvency problems, such as real estate related exposure, were independent risk factors, whose effect persisted into the latter part of the crisis. In contrast, the funding pressure on commitments-exposed banks coincided with the shifts in aggregate deposit funding. But liquidity risk interacts with solvency risk as predicted by theory. Commitments-exposed banks with weaker fundamentals were more vulnerable to the onset of the crisis than equally exposed banks with stronger fundamentals.

To conclude, the role of banks as liquidity providers was itself in crisis during the crisis from 2007 to 2008. Unlike previous crises, banks did not expand total loans and credit lines. And in the absence of a liquidity backstop by the government and the Federal Reserve, financial disruptions and business liquidations would have likely been much greater.


Elena Simintzi, Vikrant Vig, and Paolo Volpin
Labor Protection and Leverage
Review of Financial Studies | Volume 28, Issue 2 (Feb 2014), 561-591

The recent financial crisis and the subsequent global recession have once again brought the contentious topic of employment protection to the fore of the policy debate. Policymakers around the world are contemplating how to reform the rules that govern industrial relations. While some countries, such as the US, are moving towards greater labor protection, continental European countries are discussing how to amend their current labor laws in an attempt to boost potential growth. These recent developments demonstrate the need to revisit the role of labor in shaping corporate behavior and pin down the economic channel through which labor may affect firms' economic activities.

We examine the effect of labor protection on firms' capital structure decisions. Our intuition is a simple one: an increase in employment protection of labor increases restructuring costs and therefore, increases costs of financial distress at a given level of debt. Therefore, firms will now chose to hold less debt as a result of higher employment protection. This intuition follows from a simple trade-off model of capital structure in which firms trade-off the benefits of debt (debt tax shields) and the costs of debt (costs of financial distress). Pro-labor employment legislation increases the latter as labor contracts become debt-like contracts. Wages are fixed costs (i.e. operating leverage) and have to be paid every month since employees cannot be fired, resembling coupon payments of bonds. Thus, a crowding out effect is generated with operating leverage crowding out financial leverage.

Employment protection legislations

Employment protection legislations are laws which determine how difficult and costly it is to fire workers. These laws govern regular contracts, fixed-term (temporary) employment and collective dismissals, according to definitions provided by the OECD. To construct our measure of employment protection legislation that varies across countries and over time, we hand collected data on the major labor reforms across 21 OECD countries over the 1985-2007 period from a wide range of sources. In particular, we focus on changes in the procedural requirements that need to be followed when firing an employee with a regular employment contract, the notice and severance pay requirements, the prevailing standards of (and penalties for) “unfair” dismissals, the conditions under which temporary contracts can be offered, the maximum number of successive renewals and the maximum cumulated duration of these contract, the notification requirements provided by law in case of collective dismissals and the associated delays and costs for the employers.

To give an example, in 1991 a law on collective redundancies was passed in Italy. The employer has the duty to inform union representatives about the reasons for the proposed layoffs. In particular, the employer needs to show the unions that it is not possible to take alternative measures to the intended dismissals. The employer also needs to describe the measures that are planned to mitigate the social consequences of the collective dismissal. The unions may request an examination of the reasons for the layoffs and the possibilities of utilizing the workforce in different ways within the firm. If the parties fail to reach an agreement, the next step is a conciliation phase conducted by the Labor Office. Failure to follow the procedure properly is penalized by the obligation to reinstate the employees who have thus been dismissed unlawfully.

Overall, there are 21 major reforms in our sample period. Some countries have no major reform (Canada, Finland, Ireland, Japan, New Zealand, and the UK), while others have two major reforms (Austria, France, Italy, the Netherlands, Portugal, and Sweden). Nine of the reforms led to an increase in employment protection, while twelve led to a decrease in employment protection. Distinguishing between reforms that increased and those that decreased employment protection, we produced an indicator that takes values Rk,t =  + 1 (if EPL went up in country k in year t), Rk,t =  – 1 (if EPL went down in country k in year t), and Rk,t = 0 otherwise. Our indicator, EPLR, is defined recursively starting from EPL_{k,1985}^{R}=0. Thus, for any given country k in year t:  EPL_{k,t}^{R} = EPL_{k,t-1}^{R}+ R_{k,t}

It is worth clarifying that this indicator is a good indicator over time within a country and is designed to capture the long-run effects of changes in employment protection legislation. However, it is not comparable across countries, namely a higher value in one country than in another does not imply that employment is greater in the first country compared to the second one.

Worldwide evidence

To gauge the effect of employment protection legislations on leverage, we exploit intertemporal variations of employment protection legislations across countries. In our sample, US firms represent about one third of the treated sample, followed by France (with 15%) and Germany (with 11%). We compare firms' leverage in countries that are subject to a change in employment protection (henceforth treated firms) with firms in countries that have no such change (henceforth control firms).

Our empirical design, also known as difference-in-differences (DID), can be best illustrated by the following example. Suppose there are two countries, A and B, undergoing legal changes at times t=1 and t=2, respectively. Consider t=0 to be the starting period in our sample. From t=1 to t=2, country B initially serves as a control group for legal change; and after that it serves as a treated group for subsequent years. Therefore, most countries belong to both treated and control groups at different points in time. This methodology is robust to the fact that some groups might not be treated at all, or that other groups were treated prior to 1985, which is our samples start date.

We find that firms reduce their use of debt following legal changes that increase employment protection. Figure 1 presents a graphical representation of our main finding. We plot the within firm variation in leverage, as a function of changes in employment protection legislation. More specifically, Figure 1 shows the average annual leverage (measured by book value of debt over book value of total assets) in years t=-3 to t=+3 for the group of treated firms and for the comparison group of control firms, after removing the effect of firm-specific characteristics and time-varying industry conditions, which might otherwise be responsible for our results. Because of these necessary controls for a robust analysis, the annual leverage for both treated and control firms floats around zero. The treated firms have relatively higher leverage before a change in EPL and relatively lower leverage after the change, as compared to their long-run average. No such change happens to control firms. The reduction in leverage between the year before the reform and the year after is 124 basis points for a unit increase in EPL.

1: Firm Leverage around Employment Protection Legislation Changes
The figure plots the within-firm variation in leverage, as a function of changes in employment protection legislation, net of firm-specific characteristics and time-varying industry conditions. The average annual leverage for the group of treated firms (solid line) in years t=-3 to t=+3 is compared to the one for control firms (dotted line). t=0 corresponds to the year of the reform.

Our formal regression analysis leads to similar findings: the effect of an increase in EPL on total debt over assets is a reduction of 187 basis points, or 9% relative to the median leverage. The effect of an increase in EPL on long-term debt over assets is a reduction of 113 basis points, or 10% relative to the median long-term leverage.

Since the economic effect we are emphasizing in this paper goes through the labor channel, it is natural to expect that firms that have high labor turnover to be more likely affected by a change in EPL. Given these firms require higher labor turnover for their operations, increases in employment protections would impact these firms more negatively. So we divide the firms within a country according to the degree of labor turnover and we evaluate the differential effect of the EPL reform across these firms. The point of such an exercise is not to estimate the direct effect of EPL on leverage, but its differential effect across firms that differ in terms of labor turnover. As expected, we find that an increase in EPL has a greater negative effect on leverage in firms with higher labor turnover.

Our underlying assumption in this analysis, also supported by the data, is that treated and control firms are only randomly different firms. We perform a series of additional tests to make sure our findings are robust and we convincingly document that our results are not consistent with alternative interpretations. For example, we show that employment protection legislations are not sensitive to economic conditions (such as economic growth, recession, or unemployment rates). We restrict our sample of control firms to neighboring countries only. In this way, we can control for shocks that are common to a region, and our results become statistically stronger. We also control for other reforms that may be occurring at the same time as the change in employment protection and for other labor market characteristics. We also control for corporate and personal taxes, an important determinant of firm's capital structure decisions according to the literature, and the statistical significance of our results remains unchanged. Collectively, our analysis mitigates a potential concern that some unobserved variables may be driving the results.


Using firm-level data from 21 OECD countries over the 1985-2007 period, we find that leverage decreases when employment protection increases. Our intuition that can explain this finding is that pro-labor regulations, by making it difficult to adjust labor force, increase labor market rigidity. Thus, labor claims resemble debt claims, where wages can be construed as coupon payments on debt. This transformation of labor claims into debt claims crowds out firm's leverage. These findings provide supporting evidence for the critics of employment protection laws by showing that labor friendly regulation may have negative real effects via a finance channel.

Dimitris Georgarakos, Michael Haliassos and Giacomo Pasini
Household Debt and Social Interactions
Review of Financial Studies | Volume 27, Issue 5 (May 2014), 1404-1433

The role of standing relative to peers in influencing household economic behavior has been explored in many contexts, including consumption and labor supply, but less attention has been paid to how “catching up” or “keeping up” with peers (often referred to as “the Joneses”) is financed. In particular, almost no attention has been paid to whether perceptions of relative standing contribute to borrowing and to the potential for financial distress. Are people who perceive themselves as poorer than their social circle more likely to borrow and, if so, to borrow more relative to what is typically associated with their own resources and characteristics? Does such socially induced borrowing contribute to a worsening of indicators of potential financial distress, such as the debt-to-income and loan-to-value ratios?

The question of social influences on debt is distinct from that relating to consumption; concern with relative standing may lead to greater consumption, but it need not lead to a greater tendency to borrow or to run into financial distress for at least three reasons. First, households can increase labor supply, leaving room for an increase in both consumption and saving. Second, households may choose to reduce saving but may not be willing or able to raise borrowing in response to status concerns. Third, even if borrowing is undertaken, it may not significantly increase the potential for financial distress.

The Method

Investigating the influence of social interactions on debt behavior presents at least two major challenges. First, many households are willing to display their assets and consumption but prefer to leave any debts undisclosed. Thus, it makes sense to look for evidence that households adjust their debt behavior not to the debts of their peers per se but to their perception of relevant peer characteristics, like income. In the Manski terminology, instead of “endogenous effects”, one needs to focus on “exogenous (or contextual) effects”. A second challenge has to do with the scarcity of location information in household finance data. In view of privacy laws dictating anonymity of information, data collectors typically remove location details, but this step makes it impossible to identify a circle of “neighbors” or “colleagues.”

Pioneering papers that study the asset side focus on a special population group, a financial asset observed by peers, or on sociability as a factor facilitating the collection of asset-relevant information. Duflo demonstrate that individual participation of librarians in retirement investment plans is influenced by participation choices of colleagues in the same library. Hong show that more sociable individuals are more likely to own stocks.

To the best of our knowledge, our paper is the first to investigate the influence of social interactions and comparison effects on borrowing behavior. We exploit a unique feature of population-wide representative data from the Netherlands (Dutch National Bank Household Survey), namely, that the respondents report various characteristics of their peers, such as income, as these respondents actually perceive them. Thus, we can focus on whom respondents consider as peers and study their perceptions regarding their own financial position relative to them.

The Findings

We find that once we control for demographics, resources, region, time fixed effects, region-specific time trends, and other factors that typically determine borrowing needs, a higher average income in the social circle, as perceived by a household, increases a household's tendency to borrow. The estimated effects are sizeable for both collateralized and consumer debt: a 1,000 euro increase in the perceived monthly average household income of peers (corresponding to 0.85 of one standard deviation of peer income) is estimated to raise the unconditional likelihood of having collateralized (uncollateralized) loans by 10% (7%). Not only is this influence significant among those who perceive their income to be below average for their social circle, it also extends to the likelihood of future financial distress, indicated by the debt service ratio and the loan-to-value ratio. We verify the robustness of these results using several approaches, including instrumental variable estimation and placebo tests. Our aim is to rule out uninteresting alternative explanations of the peer-income-own-borrowing relation and address the potential for reverse causality or spurious correlation between the two. The former could arise, for example, if people who borrow are more likely to think of their peers as earning more than they do. The latter could be induced by similarities in unobserved characteristics with those of peers, which tend to induce both higher peer incomes and a greater tendency to borrow without any direct causal link between the two.

We also find that once the perceived peer income is controlled for, the tendency to have uncollateralize loans is partly related to the perceived spending ability of peers. Expectations about the next period's income are statistically significant but do not render peers' perceived incomes insignificant. This finding suggests that average peer income does not simply reveal prospects for the future income of the respondent (a mere “tunnel effect”) but also represents a comparison effect.

Policy Implications

Although our analysis does not rule out that much of this socially induced additional borrowing is repayable, at least ex ante, our finding that it tends to worsen indicators of potential financial distress suggests that repayment problems might still occur ex post, especially if borrowers and lenders fail to take proper account of all relevant risk factors. This paper provides a powerful additional rationale for promoting debt literacy on the part of households and scrutiny on the part of lenders: a potential for financial distress is not only generated by objective borrowing needs of households but can also be influenced by how they perceive themselves relative to their social circle. Moreover, this happens regardless of their income class and without requiring accuracy of the perceptions of relative standing.

Radhakrishnan Gopalan, Todd Milbourn, Fenghua Song, and Anjan V. Thakor
Duration of Executive Compensation
Journal of Finance | Volume 69, Issue 6 (Dec 2014), 2777-2817

We address the issue of how to measure the extent to which a particular executive compensation package is long-term in nature. One of the long-standing criticisms in the press and among politicians and regulators is that executive compensation contracts sub-optimally reward short-term performance and therefore induce myopic managerial behavior. Such rewards allegedly encourage executives to boost short-term performance at the expense of long-term value. The importance of this issue has been heightened by the fact that the apparent "short-termism" of the compensation of the CEOs of financial service firms is viewed by many as a contributing factor to the recent financial crisis. In this paper, we propose a new measure, named CEO Pay Duration, which allows us to actually quantify the extent of short-termism in executive pay. We hope our measure can help a variety of audiences. For example, boards of directors can use it to align the duration of the CEO's compensation with the strategic needs of the company. Bank regulators can use it to determine whether bank executive compensation is indeed "too short-term" in nature. Compensation consultants can use it to improve their benchmarking of executive compensation.

The measure

Our measure of executive pay duration characterizes the mix of both short-term and long-term pay. The measure is a close cousin of the duration measure commonly used in fixed income securities. We compute pay duration as the weighted average of the vesting periods of the different components of executive compensation, with the weight for each component being equal to the fraction of that component in the executive's total compensation. We calculate this measure using detailed data on the vesting schedules of restricted stock and stock options of all named executives of S&P 1500 firms during the years 2006-09.

The Questions

With this measure in hand, we empirically address the following questions:

Vesting Periods

We find that the vesting periods for both restricted stock and stock options cluster around the three to five year period, with a large proportion of the grants vesting in a fractional (or graded) manner during the vesting period. There is, however, significant cross-sectional variation in the vesting schedules and pay duration across firms. For example, executive pay duration tends to be correlated with both project and asset duration: industries with longer-duration projects, such as Defense and Utilities, offer longer-duration pay to their executives. We also find that firms in the Finance-Trading industry have above-median CEO pay duration (they rank 11th among the Fama-French 48 industries). This is somewhat surprising in light of the recent and heavy criticism that short-termism in executive compensation at banks may have contributed to the 2007-09 financial crisis. Moreover, the average pay duration increased during our sample period, especially for executives in the manufacturing and utilities industries. The average pay duration for all executives (including those below the CEO) in our sample is around 1.22 years, while CEO pay has a slightly longer duration at about 1.44 years. Executives with longer-duration contracts receive higher total compensation, but lower bonus, on average.

Empirical Findings

We find that executive pay duration is longer in firms with higher market-to-book ratio, for firms with more long-term assets, and in more R&D-intensive firms. Consistent with standard principal-agent theory, we also find that riskier firms offer shorter-duration pay contracts. We also find that firms with better recent stock performance offer longer-duration pay contracts to their executives. This may be because better returns indicate more able CEOs, and Boards of such firms may find it optimal to lengthen their compensation vesting schedules to retain the executive. Our analysis reveals an ambiguous relationship between corporate governance and executive pay duration. Some governance proxies suggest that better-governed firms use shorter pay duration, whereas other proxies suggest the opposite. Pay duration is shorter for executives in firms with a higher proportion of nonexecutive director shareholdings, for executives with higher ownership in their own firms, and in firms with less anti-takeover provisions. However, we find that it is actually longer in firms with a larger fraction of independent directors on the board.

Pay Durations

Next, we explore how pay duration is related to the incentives of the manager to manipulate short-term performance. We use the level of abnormal accruals as our main proxy for managerial manipulation of short-term performance. The use of accruals, which is part of earnings not reflected in current cash flows, accommodates a temporary shift of the firm's reported earnings between the future and the present. Firms with high abnormal accruals will have high current-period earnings and low future earnings and vice versa. We find that firms that offer shorter-duration pay contracts to their CEOs have higher abnormal accruals in the current period. This negative association is stronger for earnings-enhancing positive accruals, and is robust to controlling for known determinants of abnormal accruals. We also find that the negative association between CEO pay duration and abnormal accruals is stronger among small firms, young firms, and firms with less liquid stock.

Iftekhar Hasan, Chun-Keung (Stan) Hoi, Qiang Wu, and Hao Zhang
Beauty is in the eye of the beholder: The effect of corporate tax avoidance on the cost of bank loans
Journal of Financial Economics | Volume 113, Issue 1 (Jul 2014), 109-130

Tax savings from avoiding taxes are a real benefit to corporations. Since shareholders care deeply about after-tax profits, it is reasonable that many corporations are active and even aggressive in undertaking activities to reduce corporate taxes. Nevertheless, not all firms avoid taxes. In fact, a well-known empirical regularity is that many US firms apparently do not avail themselves of tax avoidance opportunities. Weisbach (2002) coined this phenomenon the “under-sheltering puzzle”. This study provides an answer to the “under-sheltering puzzle” by examining how debt holders, particularly banks, perceive tax avoidance activities. We argue that if debt holders perceive tax avoidance as engendering significant risks, they will price the risks into loan and bond contracts, leading to higher borrowing costs, which, in turn, moderate the incentive to engage in tax planning activities.

Why debt holders? What risks?

Debt holders are fixed claimants. They care about downside risk but face limited up-side potential. Accordingly, debt holders are naturally less sensitive to the tax savings from avoiding taxes and more sensitive to the risks engendered by corporate tax avoidance than do shareholders. Focusing on the perspective of debt holders therefore provides an empirical setting that is well-suited for the following queries. Does corporate tax avoidance engender risks? If so, do debt holders price these risks engendered by corporate tax avoidance into debt contracts? Relying on prior literature, we argue that corporate tax avoidance engenders information risk, agency risk, and IRS audit risk. The idea that corporate tax avoidance would increase the IRS audit risk is intuitive. Mills (1998) and Mills and Sansing (2000) provide early evidence on it. More recently, Desai and Dharmapala (2006, 2009) and Kim, Li, and Zhang (2011) argue and find that corporate tax avoidance engenders significant information risk and agency risk. The premise of their argument is as follows. Tax avoidance activities are necessarily complex, obfuscated, and opaque to minimize the risk of detection by the tax authorities. But if firms are hiding information from the tax authorities, they would also need to hide the same information from outside investors, shareholders, and bond holders. This peculiar characteristic of tax avoidance activities exacerbate information risk by reducing firm information quality and they give managers greater latitudes to divert corporate resources for private benefit consumption, leading to greater agency risk.

Tax-avoiding firms face higher bank loan cost

To investigate whether debt holders price risks engendered by corporate tax avoidance into debt contracts, we use a comprehensive sample of around 17,000 bank loans issued to U.S. public firms in the period 1985-2009. We focus on more aggressive tax avoidance practices because they are likely to engender greater risks. We use two book-tax difference measures (Manzon and Plesko, 2002; Frank, Lynch, and Rego, 2009) and cash effective tax rates as the main measures of aggressive tax avoidance. Across all three measures, we find a positive and significant relation between aggressive tax avoidance and loan spread after controlling for known determinants of loan spread.

We use two quasi-experiments to establish the causal effects of tax avoidance on bank loan spreads. The first involves the implementation of Financial Accounting Standards Board (FASB) Interpretation No. 48, hereafter FIN 48. FIN 48 was introduced in 2007 and it affects only those firms that undertake uncertain tax positions to avoid taxes. After FIN 48, these affected firms are required to report tax reserves related to the uncertain tax positions they have taken. However, firms with no uncertain tax position would naturally have no tax reserves to report pursuant to FIN 48. Using a difference-in-differences analysis, we find that firms that disclose a positive FIN 48 tax reserve during a three-year window immediately after FIN 48 incur significantly larger increases in loan spreads when compared to match firms that never report a positive FIN 48 tax reserve in that same period. This result indicates that banks use the tax reserves disclosed in affected firms to infer the firm's tax aggressiveness in terms of uncertain tax position.

The other experiment involves firms that are publicly scrutinized for their engagements in tax shelters. We find, in this setting, that firms affected by tax shelter news have significantly higher increases in bank loan spreads after tax shelter news when compared with match firms without tax shelter news. This result indicates that tax shelter news provides banks incremental information about the firm's tax aggressiveness, which, in turn, causes banks to increase the firm's loan spreads after its tax-sheltering activities became a publicly disclosed news event.

. . . and then some

Banks can adjust nonprice loan terms such as collateral and covenant requirement to mitigate avoidance-induced risks. Accordingly, we examine the effects of aggressive tax avoidance on the likelihood of a collateral requirement and the intensity of covenant requirement. We find that banks tighten collateral and covenant requirements when lending to firms that exhibit greater tax avoidance.

Just as banks would price avoidance-induced risks into loan contracts, so should public bond holders when issuing bonds. In fact, bond holders should be more sensitive to avoidance-induced risks than do banks because they are arms-length lenders. Using an exhaustive sample of bond issuing firms during the same sampling period 1985-2009, we find that firms with greater aggressive tax avoidance incur higher at-issue yield spreads when issuing public bonds. More important, the results indicate that tax avoidance has significantly larger incremental effects on interest spreads of bonds than on those of the bank loans, providing some evidence that public bond holders are more sensitive to avoidance-induced risks. Additionally, based on the combined sample of bond-issuing firms and loan-initiating firms over the same sampling period, we find that firms with greater tax avoidance prefer bank loans over public bonds when seeking debt financing. This latter result is consistent with Bharath, Sunder, and Sunder (2008).


Taken together, the findings on bank loan spreads, at-issue bond yields, nonprice loan terms, and firm debt financing preference for bank loans over public bonds paint a fairly consistent picture of how aggressive tax avoidance affects a firm's cost of debt capital in general. The results show that aggressive tax avoidance has multi-faceted effects on debt contracting and uniformly increases the cost of debt capital, regardless of whether the firm seeks debt financing from private bank loan market or public bond market.

These findings contribute to two under-explored issues in the tax avoidance research. First, they advance the understanding of the “under-sheltering puzzle”. Our findings suggest that, all else equal, a tax avoiding firm could incur a higher cost of debt, which, in turn, moderates the firm's incentive to engage in tax avoidance, providing a potential explanation for the “under-sheltering puzzle”. Second, Hanlon and Heitzman (2010) call for more research to explore how lenders, investors and consumers perceive corporate tax avoidance activities. Our study answers their call, and our findings show that debt holders, including banks and bond holders, view corporate tax avoidance negatively as they perceive tax avoidance as engendering significant risks.


E. Han Kim and Paige Ouimet
Broad-Based Employee Stock Ownership: Motives and Outcomes
Journal of Finance | Volume 69, Issue 3 (Jun 2014), 1273-1320

Firms initiating broad-based employee stock ownership plans often claim ESOPs increase worker productivity by improving employees' personal incentives and team work. But skeptics are hesitant to buy into this reasoning. For one, free-riding may negate incentive effects; in other words, non-executive level employees may feel they have little impact on the stock price and therefore be unwilling to alter their behavior in tasks requiring additional effort or sacrifice. For another, the real motive might be to conserve cash by issuing stocks to employees in return for lower wages, or to thwart hostile takeover bids. ESOPs established for such purposes are unlikely to improve worker productivity. These concerns call for sorting out different motives, which we do by separating ESOPs into small and large ESOPs. Small ESOPs are defined as those never controlling more than 5% of the firm's outstanding shares. Firms would not implement these small ESOPs to conserve cash or prevent hostile takeover threats; to achieve those goals requires much larger scale ESOPs. We also separate ESOPs by the likely intensity of the free rider problem. Free riding is likely to be less severe when there are fewer employees, so we separate firms into not-so-numerous- versus numerous-employee firms.

With these stratifications, we begin our investigation by asking: If an ESOP is established for the sole purpose of improving employee incentives, and if the free rider problem is not severe, does it increase worker productivity? How are employees affected financially? How about stockholders?

Small ESOPS, not many employees

We answer these questions by examining small ESOPs adopted by public firms with not-so-numerous-employees, which are most likely to be motivated by the incentive purpose and less likely to suffer from free riding. Specifically, we estimate how adopting these ESOPs affect employee compensation, shareholder value, and worker productivity. We also investigate how productivity gains, if any, are shared by employees and shareholders. Wage and employment analyses are conducted with the U.S. Census Bureau database, which provides micro data on employee payroll at the establishment level. An establishment is a workplace, such as a factory, an office, a research lab, a restaurant, and so on. We find that small ESOP adoptions by not-so-numerous-employee firms do increase employee compensation. On average, cash wages—all forms of taxable ordinary income, such as regular paychecks, bonuses, and commissions—increase by 20% following these ESOP adoptions, relative to a control group of non-ESOP establishments. These cash wages do not include the value of ESOP shares granted; hence, our estimates underestimate real increases in total employee compensation. Shareholder value also increases, on average, by 21% relative to industry peers without ESOPs. Since employees and shareholders are the two main claimants to firm surplus, these estimates clearly imply substantial productivity gains. This inference is corroborated by direct estimates of changes in total factor productivity (TFP) for manufacturing firms that adopt small ESOPs with not-so-numerous employees; their TFP increases significantly. In addition, these ESOP adoptions are followed by increases in both employment and the number of establishments, indicating more hiring and more investment.

Gains depend on bargaining power

How are the productivity gains shared by employees and shareholders? It depends on the employees' bargaining power relative to the employers. We measure employee bargaining power by employer concentration within the industry and geographic location of each workplace, a measure akin to estimating alternative employment opportunities in the same industry and location. (This measure assumes switching to a different industry and relocating to a different region are costly to workers.). We find that when employee bargaining power is weaker, wage gains are smaller and shareholder value gains are greater, and vice versa. These results are not driven by pre-ESOP growth opportunities and firm performance, nor are they driven by anticipated changes in worker bargaining power at the time of ESOP adoption.

Large ESOPs have small gains in production

Our investigation also includes large ESOPs controlling more than 5% of the firm's outstanding shares at any point in time. When not-so-numerous-employee firms adopt them, there are no noticeable effects on cash wages, shareholder value, or employment. However, cash wages do not include the value of ESOP shares granted to employees. The average market value of shares granted through these large ESOPs in our sample is $26,796 (in 2006 dollars) per employee, equal to 5.06% of annual wages if the shares were allocated equally over 10 years. When this value is taken into account, large ESOPs seem to increase total employee compensation. Our estimates of total factor productivity for manufacturing firms also indicate an increase in productivity. However, the overall evidence suggests that large ESOPs at not-so-numerous-employee firms are associated with considerably smaller productivity gains than a small ESOP. The smaller gains are due to the different motives behind the adoption of large ESOPs.

Sell shares in the open markets to raise cash

Some large ESOPs are implemented by cash-constrained firms to conserve cash by substituting cash wages with ESOP shares. Employees cannot sell these shares until they leave the company or are close to retirement age. Because of the sales restrictions, employees are exposed to risk that can be diversified by other investors. Risk-averse employees will therefore value ESOP shares below market price. Thus, if the purpose is to raise cash, it would be better to issue shares in the open market than issuing shares to employees through ESOPs. However, cash-constrained firms often have limited access to external financing, and some of them resort to issuing shares to employees through large ESOPs.

ESOPs as an antitakeover device hurt shareholders

Another motive unrelated to improving employee incentives is for management to form an alliance with workers to garner their support in thwarting hostile takeover threats. A number of states have business combination statutes (BCSs) that impose a temporary moratorium on takeover bids if a sufficiently large block of shareholders unaffiliated with management, such as a large ESOP, vote against the takeover bids. If a firm's state of incorporation has such a BCS, the firm can adopt a large ESOP as an antitakeover device. This scheme requires employee support, which the management may buy with higher wages. Cash wages indeed increase following the adoption of such ESOPs. Thus, when management uses ESOPs for the entrenchment purpose, it costs shareholders in two ways: unearned wage increases and the foregone takeover premium.

With few employees, small ESOPs enhance incentives

When firms with numerous employees adopt ESOPs, the effect on cash wages and shareholder value is more or less neutral, which is attributable to the free rider problem. There are, however, a few large ESOPs increasing both cash wages and worker productivity, indicating some large firms are able to overcome free-rider effects by giving employees bigger shares of the firm.

To summarize, employee capitalism works when the true intent is to enhance incentives for a moderate number of employees. These ESOPs can improve team work and encourage co-monitoring among workers, leading to significant increases in productivity. The productivity gains are shared by employees and shareholders according to their relative bargaining power. Average wages, shareholder value, and the level of employment all increase. These ESOPs are win-win plans. The same cannot be said when the number of employees is numerous and/or the size of the ESOP is large. The free rider problem and/or non-incentive purposes such as cash conservation and management entrenchment negate much of incentive effects that could arise with employee share ownership.


Jongha Lim, Bernadette A. Minton, and Michael S. Weisbach
Syndicated Loan Spreads and the Composition of the Syndicate
Journal of Financial Economics | Volume 111, Issue 1 (Jan 2014), 45-69

During the past decade, non-bank institutional investors are increasingly taking large roles in the corporate lending, especially in the syndicated loan market. A puzzling aspect of this phenomenon is that non-bank institutional investors, especially hedge funds and private equity funds, have substantially higher required rates of return than banks, yet participate in the same loans and consequently receive the same returns on these investments.

Why do some facilities have participation of non-bank investors while others do not? Why do various types of institutional investors, whose required ex ante returns are substantially different, invest in the same loan facilities as one another? How does their presence affect the loan spread? In this paper, we address these issues using a sample of 20,031 facilities (or tranches) of leveraged loans that were originated between 1997 and 2007.

1: Relative Portion of “Non-bank” versus “Bank-only” facilities
There is substantial participation by non-bank institutions.

We first document that there is substantial participation by non-bank institutions in these loans. Of the 20,031 leveraged loan facilities, 13,752 are associated with a syndicate containing only commercial or investment banks (bank-only facilities), while the remaining 6,279 have syndicates containing at least one non-bank institutional investor (non-bank facilities) (see Figure 1). These institutional investors are most often finance companies (4,603 loan facilities), private equity or hedge funds (2,754 loan facilities), and mutual funds (1,010 loan facilities).

Non-bank facilities have higher spreads than bank-only facilities, controlling for other factors that affect the loan facility's spread such as the firm's risk (measured by either firm-level accounting variables or the firm's credit rating), the loan facility's type (Term loan A, Term loan B, or revolver), and other facility-specific characteristics (maturity, loan size, security, etc.). The non-bank premium ranges from 21 to 76 basis points, depending on how we control for the borrowing firm's risk and other differences between loan facilities (see Figure 1). All these estimates are, however, statistically significant.

Hedge funds and private equity firms earn a premium

Furthermore, we find that the size of non-bank premium varies by the type of non-bank syndicate member and that, not surprisingly, the effect is particularly large when private equity or hedge funds are part of the syndicate. In contrast, other types of non-banks institutional investors such as insurance companies or mutual funds receive essentially no abnormal premium.

1: Average non-bank premium (basis points)
Sample Average spread Non-bank premium % of total spread
By facility type
All facility types 256.6 53.9 21.0%
Revolvers 230.1 44.9 19.5%
All, term loans 301.2 68.9 22.9%
Term, loan B only 305.8 74.5 24.4%
By issuer credit rating
All, with ratings 237.6 22.2 9.3%
BBB, and above 151.4 28.6 18.9%
BB 201.7 20.8 10.3%
B, and below 298.9 21.1 7.1%
Unrated 270.4 76.0 28.1%
Non-bank facilities have higher spreads with premiums ranging from 21 to 76 basis points.

An important consideration is the extent to which the differences between bank and non-bank tranches reflect differences in risk between these tranches. It is likely that non-bank investors tend to invest in riskier borrowers than do banks. The situation is more complex in a syndicated loan in which banks and non-banks invest side-by-side. Nonetheless, it is possible that loans containing non-bank investors could be riskier on average than loans without them.

Premiums do not reflect risks

To evaluate whether differences in risk can explain the non-bank premium, we use an approach that relies on comparison of different facilities within the same loan package. Different facilities within the same loan package are issued by the same firm, being originated at the same time and typically having the same seniority and covenants. Therefore, default risk and creditor rights attached to them are essentially the same. Then, once other facility-specific characteristics are controlled for, the incremental effect of a non-bank participant on the relative pricing gap between facilities within a given loan should reflect the impact of non-bank syndicate participation rather than underlying risk differences.

Premiums are high when capital constraints are tight

Our results suggest that the presence of a non-bank investor in a facility is associated with a higher spread, even measured relative to other facilities of the same loan that have the same default risk. When a loan package has both a revolver tranche and a Term loan B tranche and the non-bank member is present in the Term loan B portion, the spread gap between the two types of tranches is 33 to 42 basis points higher than it otherwise would be if both tranches were funded by banks only. The incremental impact of non-bank participation on the spread gap between Term loan B and Term loan A facilities ranges between 8 and 10 basis points, although such effect is not statistically significant.

...because borrowers are capital constrained

We interpret this result as consistent with the view that the non-bank institutions are providing capital when the lead bank has difficulty filling the syndicate solely from banks. The lead bank is likely to have difficulty filling the syndicate solely from banks when the borrowing firm is financially constrained and therefore traditional lenders are reluctant to extend credit to the firm. Financial constraints facing the borrowing firm would further make alternative options of financing more expensive. Using various measures of financial constraints, such as whether the firm has credit rating, whether the firm has accessed public bond market, the size of debt coming immediately due, and the 'size-age' index, we find that each measure of financial constraint is associated with significantly higher non-bank premium.This result suggests that non-bank investors can come in to fill the financing void left unfilled by traditional lenders and receive premium in return. This result suggests that non-bank investors can come in to fill the financing void left unfilled by traditional lenders and receive premium in return.

...or when supply of credit is tight

Similarly, if non-bank premiums reflect a return to providing capital at times when banks cannot, then the premiums should vary depending on the supply of bank capital available at a particular point in time. Factors that could affect the supply of bank capital include the risk aversion of banks, the overall state of the economy, as well as the demand for loans from collateralized loan obligations (CLOs). Thus, we investigate the way non-bank premiums vary year by year. Although there is significant non-bank premium in each year of our sample, there is substantial inter-temporal variation in non-bank premiums, with the premium ranging from 35 basis points in 2005 to 69 basis points in 1997 (see Figure 2). Moreover, consistent with the notion that these non-bank premiums reflect bank risk preferences, we find inter-temporal variation in non-bank premiums is closely correlated with the high-yield credit spread, which represent the compensation the market provides for holding riskier securities (and thus the degree of risk aversion of market participants).

To test whether non-bank premiums are related to the high-yield spreads more formally, we include an interaction term between the high-yield spread and the non-bank participation indicator in a regression estimating the loan spreads. We find the coefficient on this interaction term is significantly positive, implying that non-bank institutions earn a higher return for providing capital at times when banks are most reluctant to do so.

2: Inter-temporal variation in non-bank premiums and high-yield spread
There is substantial inter-temporal variation in non-bank premiums.

Another implication of the view that the non-bank premiums reflect the provision of liquidity by the non-bank institutions is that the premiums should be higher when the lead bank in the syndicate has limited liquidity itself. Consistently, we find evidence that non-bank premiums are particularly large when the lead bank lacks liquidity, measured by the lead's cash to total assets ratio or by the indicator for whether the lead is active on securitization or not.

Overall, the evidence in this paper suggests that when a lead arranger of a syndicated loan is unable to raise entire capital solely from fellow banks, it sometimes approaches non-bank institutions to help filling the syndicate. To attract these non-bank investors, the lead bank offers a higher spread. Non-bank premiums in syndicated loans appear to represent compensation non-bank investors receive for providing capital at times when filling the loan is relatively difficult.

As debt markets mature, it seems evident that non-traditional players will provide capital to a larger degree than has been true historically. Why is it optimal to have different types of investors providing the capital for the same loans? To what extent does borrower performance depend on the provider of capital? Is there important variation across banks that lead some to be more prone to co-invest with hedge funds and private equity funds in loans with higher spreads? Understanding the answers to these and related questions would be a useful direction for future research.


Wai-Man Liu and Phong Ngo
Elections, political competition and failure
Journal of Financial Economics | Volume 112, Issue 2 (May 2014), 251-268
In Washington, the view is that the banks are to be regulated, and my view is that Washington and the regulators are there to serve the banks
Rep. Spencer Bachus — chairman of the House Financial Services Committee in the 112th Congress, The Birmingham News on 8 December 2010.


The relationship between banking and politics is an intimate one. Governments control the supply of banks in the economy through chartering restrictions and licensing, they set up institutions that provide depositors with insurance and banks with a lender of last resort, and routinely set rules that attempt to govern the risk taking behaviour of banks.

Indeed, the bi-annual “Banking Banana Skins” survey by Pricewaterhouse Coopers and the Centre for the Study of Financial Innovation finds that “political interference” was rated as the number one risk that banks faced in 2010. This is surprising given the international banking system had witnessed possibly the worst crisis on record which was largely attributed to credit and liquidity risks, and at the same time, ironic, given the banks were bailed out by politicians using public money.

This active role of government in the banking sector creates an incentive problem. On the one hand, government play a role in the creation of institutions that make a banking system possible. On the other hand they quite often look to the banking system to facilitate their own political survival. Political support can be indirect, through say, subsidized lending to preferred industries or direct in the form of campaign contributions or a share of profits due to ownership. For example, according to the Centre for Responsive Politics, Spencer Bachus, who is the chairman of the House Financial Services Committee in the 112th Congress, raised over $2.3 million in campaign funds in 2011-2012 with the top five industries being commercial banks, securities and investment, insurance, real estate and finance/credit companies contributing over 40%.

So while a healthy banking system can be huge source of benefit for politicians, bank failure on the other hand, can get politicians into electoral hot water. Politicians therefore have incentives to interfere with bank closure rules to, for example, favour preferred (politically connected) constituents or simply to avoid the political costs associated with failure.

Examples of Political Interference

There have been several examples in the popular press of political interference in the banking system. Probably the most famous case is that of Lincoln Savings and Loans, in which five US senators (known as the “Keating Five”) were accused of improperly intervening in a regulatory investigation of Charles H. Keating, Jr. (Chairman of the Lincoln Savings and Loan Association) by the Federal Home Loan Bank Board (FHLBB) in 1987. Lincoln Savings and Loans eventually collapsed in 1989, at a cost of over $3 billion to the federal government. The substantial political contributions Keating had made to each of the senators, totalling $1.3 million, attracted considerable public and media attention. This lead to a Senate Ethics Committee investigation in which three of the senators were found to have “substantially and improperly interfered with the FHLBB's investigation” and the other two, while being cleared, were criticized for exercising “poor judgement”. All five senators served out their terms however only the two ran for re-election.

A more recent example is that of Cleveland thrift AmTrust, whose failure was delayed by 11 months because Ohio Congressman Steven LaTourette and Cleveland mayor Frank Jackson intervened when the Federal Deposit Insurance Corporation (FDIC) tried to seize and sell the institution in January 2009. AmTrust was issued with a cease and desist order in November 2008, and when it failed to recapitalise by the deadline of December 31, 2008 the FDIC stepped in. The local politicians were able to delay the failure by convincing Treasury and the White House to keep the FDIC at bay. By the time AmTrust was finally seized by the FDIC on December 4, 2009 its common equity had fallen by $667 million to $276 million from the year before. The failure cost the FDIC insurance fund $2 billion.

Is political interference systematic?

Are these incidents isolated cases? Just a few “bad eggs”—or are they representative of a more systematic phenomenon? A natural place to look for systematic evidence of political interference in banking is around elections as this is when bank failure can potentially be the most costly to a politician. Bank failure typically leads to costs that are borne by the taxpayer (for example, due to losses to the insurance fund), leading the electorate to question the competency of the incumbent in regulating the banking sector. Accordingly, politicians have the incentive to take costly action to delay bank failure in election years. The economic cost of delay (possibly from larger losses to the insurance fund than would otherwise be the case) is widespread across tax payers, whereas the benefits are concentrated with interest groups like bank owners, employees and uninsured depositors—which further exacerbate the political incentive to delay bank failure in an election year.

Bank failures probability before elections

Using data from the United States between 1934 and 2012, covering all failed banks (3995) documented by the FDIC, we study the timing of bank failure around gubernatorial elections (state elections for the office of governor). In the figure below, the blue bars plot the number of bank failures in 3-month blocks leading up to an election while the pink bars plot the number of bank failures in 3-month blocks in the months after state elections. The blue and pink horizontal dashed lines represent the average number of bank failure in a 3-month period before and after elections respectively.

While Figure 1a shows no discernible difference between the pre- and post-election failure rates (235 vs. 240 failures per 3-months respectively), a striking picture emerges when we control for the clustering of bank failure around crises: bank failure is much less likely in the months leading up to an election than in the months after. In fact, based on these raw numbers alone, bank failure is about 40% less likely in the months leading up to an election compared to the months following an election-the pre- and post-election average number of bank failures are 59 and 97 failures per three-months respectively.

1: Bank Failure around Gubernatorial Elections


The figure below plots the frequency of bank failures around gubernatorial elections for all recorded failures (3995) of FDIC insured financial institutions (commercial banks and thrifts) between 1934 and 2012. Panel (a) plots bank failures in all periods. Panel (b) plots banks fails in non-crisis periods only, i.e. failures occurring during the S&L Crisis (1986-1992) and the Global Financial Crisis (2007-2010) are excluded.

In Figure 1, we have used the term `bank failure' to mean both outright failures in which banks lose their charter and cease to operate as well as FDIC assistance transactions in which a failing institution is restructured with FDIC assistance and allowed to continue to operate under its existing charter. To examine whether political concerns determine the type of failure transaction chosen by regulators around elections we split all failures into two sub samples depending on the failure type. In Figure 2 we plot the frequency of bank failures around elections separately for outright failures (Figure 2a) and assistance transactions (Figure 2b).

2: Election cycles and bank failure - by type of failure


The figure below plots the frequency of bank failures around gubernatorial elections for all recorded failures of FDIC insured financial institutions (commercial banks and thrifts) between 1934 and 2012 in non-crisis periods only, i.e. failures occurring during the S&L Crisis (1986-1992) and the Global Financial Crisis (2007-2010) are excluded. Panel (a) plots the frequency of outright bank failures. Panel (b) plots bank failures that are classified as assistance transactions.

Like Figure 1, the blue bars plot the number of failures in 3-month blocks leading up to an election while the pink bars plot the number of failures in 3-month blocks in the months after elections. The blue and pink horizontal dashed lines represent the average number of bank failures in a 3-month period before and after elections respectively. There is a remarkable difference between the two figures: outright failures are clearly less frequent prior to elections whereas assistance is much more frequent-politicians tend to bailout banks or delay the failure of banks due to electoral concerns.

Keeping bankrupt banks open increases losses

So what can be learned from this? First, political interference in efficient bank failure can lead to larger losses than would have otherwise been the case: allowing banks with zero or negative net worth to continue to operate allows them to “gamble for resurrection” by taking on excessive risk. Second, if banks know that they are likely to be bailed out due to political concerns, this might encourage them to take on more risk-the standard moral hazard problem, though now it does not only apply to banks that are “too big to fail”, but potentially any bank if it is an election year. Both imply that ex-post bank closure rules play an important role in determining ex-ante bank risk taking.

However, since the global financial crisis, regulators around the world have been “fixing” bank regulation by introducing new capital and liquidity requirements for banks while, for the most part, ignoring the important role of bank closure policy.

Regulators should be independent

In closing, we have some good news and some bad news:


Avner Kalay, Oguzhan Karakas, and Shagun Pant
The market value of corporate votes: theory and evidence from option prices
Journal of Finance | Volume 69, Issue 3 (Jun 2014), 1235-1271

What is the value of a corporate vote? Since the seminal work of Berle-Means (1932), it has been widely held that one of the most important contractual rights that shareholders have is the right to vote in corporate elections. Estimating the value of the voting rights embedded in stocks is important to our understanding of corporate control, but is not trivial. In our paper, "The Market Value of Corporate Votes: Theory and Evidence from Option Prices" (Kalay-Karakas-Pant (JF 2014)), we propose a new method to measure the market value of shareholder voting rights.

We quantify the value of voting rights as the difference in the prices of the stock and the corresponding synthetic stock. The synthetic (non-voting) stock is constructed using option prices, particularly facilitating the put-call parity relationship. The main insight is that the owners of common stock have both cash flow rights and voting rights, whereas the holders of synthetic stock have the cash flow rights but not the voting rights. Hence, the difference in the price of the stock and the synthetic stock quantifies the value of voting rights during the expected life of the synthetic stock.

Credit risk increases after CDS introduction

Evidence on the value of voting rights thus far has been focused on three methods of estimation. First, one can compare the prices of multiple classes of shares having identical cash flow rights and differential voting rights. Second, one can consider sales of controlling blocks in publicly traded companies and compare the price paid per share for the block with the prevailing stock price right after the transaction. Third, one can quantify the value of the vote as the incremental cost of borrowing stock (implied by the equity lending fee) around the record dates related to shareholder meetings.

All approaches have advantages but also their problems. For instance, the first method requires that at least two classes of shares be publicly traded. However, only a small percent of US publicly traded companies are dual-class firms. Moreover, the two classes usually differ in their liquidity. More importantly, these samples are potentially subject to selection biases. With the second method, it is not possible to measure the value of control when the controlling block is not transferred. The potential selection bias and the small sample size are other limitations. Finally, for the third method, given that the equity lending market is negotiated, decentralized, and opaque, one cannot be certain if the reported lending rate is in fact the clearing rate in the market. Additionally, the value of the vote inferred from prices in the equity lending market can be significantly downward biased. With our technique, one can estimate the market value of voting rights attached to a stock, as long as the stock has call and put option pairs with the same maturity and strike price traded. Thus, our method can be used to estimate the voting premium for a large cross section of stocks at any point in time.

Voting rights are worth 2% of stock price

Empirically, using our measure of the voting rights for 4,768 public firms in the US over 1996-2007, we estimate the mean annualized value of a voting right to be 1.58% of the underlying stock price. This compares to 1-10% for the value of vote estimate with other methods for US public firms. Zingales (QJE-1995) notes “...the price of a vote is determined by the expected additional payment vote holders will receive for their votes in case of a control contest.” This would imply that the value of voting rights depends on both the probability that a voting event occurs and the economic significance of such a voting event. As such, the value of voting rights is expected to be time-varying. Consistent with this theory, we find the value of voting rights to increase around shareholder meetings that are likely to be more contentious - such as special meetings (see Figure 1), meetings with a high-ranking agenda (e.g., antitakeover, mergers & reorganizations), and meetings with ex-post close votes.

1: Value of voting rights around shareholder meetings
The value of voting rights increases around contentious meetings.

In addition to shareholder meetings, we analyze episodes of hedge fund activism and M&A events. We find that the value of voting rights substantially increases for firms targeted by activist hedge funds for the period 1998 to 2008. The increase in the value of voting rights is higher for hostile engagements compared to non-hostile ones. These results are important as they demonstrate that the mere threat of a voting event is sufficient to increase the value of voting rights.

Studying the M&A events over the period 1997 to 2005, we find a substantial jump in the value of voting rights on the announcement date of an M&A event. We observe a significant drop in the value of the voting rights at the merger completion date. We document a large drop only for deals that were effective (i.e., not withdrawn).

The documented decrease in the value of shareholder voting rights at the completion of an M&A deal demonstrates that it can be optimal to exercise deep-in-the-money call options prior to expiration, even if the underlying stock pays no dividends. This might help explain some of the early exercise puzzles in the literature. To the best of our knowledge, this study is the first to point out that early exercise of call options can be optimal even in the absence of dividends on the underlying. In a similar fashion, some put option holders will find it optimal to delay exercise until after the drop in the value of the vote.

Finally, we show that the value of voting rights is not bounded by exogenous arbitrage activity - to the contrary - the value of voting rights is an important ingredient in the cost of the put-call arbitrage activity. We present evidence suggesting that transaction and shorting costs unrelated to the vote do not affect our findings.


Our paper contributes to the literature on corporate governance by introducing a new method using option prices to estimate the value of shareholder voting rights. The method can be used to study the determinants of the voting premium. Indeed, the method can be applied to any study focusing on corporate control. The method might also be useful for understanding control related issues in the financial regulation framework. Finally, the method has implications for the asset pricing literature as it highlights the importance of the value of voting rights in put-call parity violations. This literature does not typically account for the value of voting rights.


Kyonghee Kim, Elaine Mauldin, and Sukesh Patro
Outside Directors and Board Advising and Monitoring Performance
Journal of Accounting and Economics | Volume 57, Issues 2-3 (Apr 2014), 110-131

The effort and performance of directors on the boards of public corporations are generally examined from the twin perspectives of monitoring and advising company management. Recent research on corporate boards examines whether the board's effectiveness at monitoring tasks comes at the cost of its performance in advising tasks and vice-versa. While some recent empirical evidence supports such a view (Faleye-Hoitash-Hoitash JFE 2011), a commonly held opposing view is that the incremental efforts made in carrying out one type of task can have positive spillover effects in performing the other type of task. For example, to advise the CEO on optimum dividend policy the director needs to develop a proper understanding of the firm's investment opportunity set, but such understanding is likely to help the board in better monitoring the CEO's performance and compensation (Brickley-Zimmerman JAE 2010). In Kim-Mauldin-Patro (JAE 2014), we contribute to this discussion by examining whether the performance of the board in fulfilling its monitoring duties is achieved at the cost of performance in advising tasks.

A distinct focus on the advising versus the monitoring performance of the board in governance research not only derives from a separate interest in each but also from the existing measures of board structure. Outside directors contribute primarily to the monitoring function because they are independent of management and inside directors contribute primarily to the advising function because they have more firm-specific knowledge (Lehn-Patro-Zhao FM 2009; Linck-Netter-Yang JFE 2008). Constraints on board size imply that greater monitoring strength accruing from board independence will come at the cost of reducing the board's advising capacity. We focus on an aspect of board structure that has received relatively scant attention in the empirical literature on boards and that is likely to capture both advising and monitoring capabilities of outside directors - their tenure. We focus on tenure because the central challenge for outside directors in performing both monitoring and advising roles is to overcome problems arising from information asymmetry between themselves and firm management. The firm-specific and management-specific knowledge that can overcome these problems is likely to increase with the tenure of outside directors through a variety of avenues including more board and committee meetings attended, likely increased committee assignments, greater experience with the firm's strategies and policies, and greater within-firm deal-level experience. Because such knowledge acquisition can help reduce information asymmetry and attenuate informational conflict between the board and management, we hypothesize that outside director tenure will have a positive impact on the performance of the board.

While the above arguments are applicable for most advising tasks and monitoring tasks, there are certain specialized monitoring tasks where they may not apply. Specifically, the specialized monitoring task of ensuring the quality of financial reporting requires particular functional expertise and experience, requirements that are not met through increased firm- and management-specific knowledge alone. Because efficacy in financial reporting monitoring demands a high level of technical detail and knowledge of accounting standards and concepts, internal control concepts, and auditing processes, it is likely that the financial expertise of outside directors, rather than their tenure, has a positive impact on this monitoring task. Thus, we hypothesize that the tenure of outside directors has a positive impact on advising tasks and generalized monitoring tasks such as compensation monitoring but not on specialized monitoring tasks such as financial reporting monitoring.

Data and Design

We use detailed Morningstar data on directors for a sample of about 14,000 firm-years from 2003 to 2008 to test our hypotheses. Because tenure is likely affected in mechanical ways or in ways that are not reflective of gaining firm-specific knowledge, we use the residual from the following regression as our primary independent variable (Outside Tenure):   text{Raw Outside Tenure}_t =  alpha +  beta_1  text{Inside Tenure}_t+ beta_ 2  text{Firm Age}_t : - :1 :     + :  beta_3 :  text{IRisk}_{t,t-2}+ beta_4  text{Altman Z}_{t-1}+ beta_5  text{ROA}_{t-1}+ epsilon Inside Tenure, the average tenure of inside directors, accounts for director turnover associated with changes in top management (Hermalin-Weisbach RJE 1988). Firm Age, accounts for mechanically shorter director tenure in younger firms. Idiosyncratic risk (IRisk) captures the impact of firm-specific volatility on director tenure. We expect greater volatility leads to higher levels of director turnover and shorter tenures. The modified Altman's Z-Score, accounts for board changes associated with weak financial performance (Hermalin-Weisbach RJE 1988). Because Altman-Z may not fully capture the impact of firm performance, we also include the firm's return on assets (ROA) to incrementally capture performance. To properly isolate the impact of tenure, we include a comprehensive set of other board characteristics in all our analyses. These include, the supervisory and financial expertise of outside directors, a measure of the skill of outside directors (based on the number of other directorships held and tenure in these positions), and board size and independence.

Outside Director Tenure and Board Advising

For advising performance, we examine the acquisition and investment decisions of the firm because these are characterized as typical advisory functions of the board. While acquisitions may reflect the firm's need for growth and strategic recombination, a substantial part of the literature on acquisitions examines other causative factors arising from agency motivations and managerial hubris. Under this view, a manager who wants to maximize compensation is likely to make acquisitions more frequently and/or undertake larger deals. If boards are able to differentiate value-increasing acquisitions from those merely aimed at increasing firm-size, higher advising performance will be reflected in fewer and smaller acquisitions, and in deals with positive announcement returns and post-acquisition performance (Gompers-Ishii-Metrick QJE 2003). As Table 1 shows, we find that Outside Tenure is negatively associated with the likelihood, frequency and relative size of acquisitions, and positively associated with acquisition announcement returns. We find weaker evidence that tenure has a positive impact on post-acquisition returns (untabulated). In robustness tests, we find no support for the potential alternative explanation that firms bring on new directors in anticipation of acquisitions. Table 1 also shows that longer outside director tenure is associated with smaller amounts of over- or under-investment. Overall, these results suggest that firms are more selective when making acquisition and investment decisions as outside director tenure increases.

1: Outside Director Tenure and Board Advising
Acquisition Acquisition Acquisition Announcement
Likelihood Frequency Ratio Return Abnormal
Outside Tenure (t) –0.032** –0.060** –0.033** 0.003** –0.067**
Outside Skill (t) –0.016 0.108** 0.035*** 0.000 –0.405**
Inside Tenure (t) –0.007** –0.007 –0.008** 0.000 –0.042**
Outside Supervisory (t) 0.066 –0.022 –0.131*** –0.021*** 0.045
Outside Financial (t) 0.227*** 0.250 –0.294** –0.024*** –0.367
Board Independence (t) 0.140 0.148 –0.186 0.021 –0.387
Board Size (t) –0.017 0.011 –0.006 –0.000 0.011
Inverse Mills Ratio - - 0.685** –0.004 -
Constant t –2.302** –3.032** –0.841** 0.021 8.264**
Year, Industry, Control Variables are included
Observations 13,736 13,736 951 951 11,750
Non-zero obse 951 951 951 951
Pseudo/Adj. R2 0.0749 (p<0.01) (p<0.01) (p<0.01) 0.1279

Outside Director Tenure and Board Monitoring

For monitoring performance we examine CEO compensation monitoring and financial reporting monitoring. Table 2 shows that excess CEO compensation, i.e., after accounting for the impact of normal economic determinants, decreases as Outside Tenure increases, suggesting that outside directors' knowledge about the firm and the manager acquired over time enables them to limit managerial rents. For financial reporting

Finally, we use the firm's operating performance as a summary measure of board performance in both advising and monitoring. Pooled OLS regressions show that Outside Tenure is positively associated with operating performance measured by annual industry-adjusted return on assets, return on equity, net profit margin and asset turnover ratio measured in the subsequent year. Because of a potential reverse causality issue wherein outside directors leave firms that have impending performance declines, we use an instrumental variable (IV) approach to re-estimate the relation between director tenure and firm performance. We use outside director age as the instrumental variable. Director age and director tenure are highly correlated, not only mechanically but also because general labor market trends suggest that older workers are less likely to change their job. Importantly, research finds little or no association between age and performance in a meta-analysis of 380 empirical studies, and a small negative association between age and executive functions across 34 empirical studies (Ng-Feldman JAP 2008; Rhodes PA 2004). Related to directors, Ferris-Jagannathan-Pritchard (JF 2003) suggest that any positive effects from director experience increasing with age may be offset by older directors having less energy, posing a last period risk, and viewing directorships as lucrative part-time jobs for their retirement years. Consistent with the OLS results, we find that IV estimation yields a positive and significant association between outside director tenure and firm performance (not tabulated here).

2: Outside Director Tenure and Board Monitoring
Excess, CEO Comp (t+1) Discretionary,Accruals (t+1)
Outside Tenure (t) –0.010** 0.058***
Outside Skill (t) 0.014 0.155
Inside Tenure (t) –0.005** 0.050**
Outside Supervisory (t) 0.012 –0.835
Outside Financial (t) –0.051 –1.418**
Board Independence (t) 0.325** 0.356
Board Size (t) –0.007
Constant –0.237*** 3.799**
Year Effects Yes Yes
Control Variables Yes Yes
Observations 10,295 12,638
Adjusted R2 0.0221 0.0437


Throughout our analysis, we test for the potential non-linear impact of director tenure (Vafeas JBFA 2003) but find no significant evidence. Overall, our results make two contributions. First, they show that outside director tenure, a board characteristic that has heretofore received scant empirical attention, is an important dimension of the advising and monitoring capability of the independent board. Second, they provide additional insights on the dual roles of the board, advising and monitoring, and thus add to recent work that emphasizes distinctions and competition between the two roles (Faleye-Hoitash-Hoitash JFE 2011). While our financial reporting monitoring results suggest some support for the

Rainer Jankowitsch, Florian Nagler and Marti G. Subrahmanyam
The Determinants of Recovery Rates in the US Corporate Bond Market
Journal of Financial Economics | Volume 114, Issue 1 (Oct 2014), 155-177

The recovery rate in the event of default is an important risk factor in pricing financial contracts exposed to credit risk. More recently, the occurrence of several default events has highlighted the stochastic nature of recovery rates for corporate bonds. It is, therefore, important to understand the determinants of this risk factor in greater detail. In this paper, we examine the recovery rates of defaulted US corporate bonds, based on a complete set of transaction data over the time period from 2002 to 2010 obtained from the Trade Reporting and Compliance Engine (TRACE), a database maintained by the Financial Industry Regulatory Authority (FINRA). This data source is rather unique for an over-the-counter (OTC) market since, in almost all other cases, price information must typically be obtained either from an individual dealer's trading book, which provides a very limited view of the market, or by using bid-ask quotations, which are often unreliable in OTC markets. In particular, we investigate the trading microstructure of defaulted bonds for a broad set of default event types covering formal bankruptcy procedures, out-of-court restructurings and downgrades to default status by rating agencies representing payment defaults and unlikely-to-pay events. This analysis allows us to provide reliable market-based estimates of the recovery rates, study liquidity and, hence, study price formation for individual defaulted bonds.

Market-based recovery rates are highly relevant for investors

The global financial crisis has highlighted the importance of credit risk in the pricing of financial contracts and emphasized the multifaceted nature of its key determinants: the probability of default and the recovery rate in the event of default. Traditionally, credit risk modeling has focused on the probability of default, while the recovery rate has been set to parametric values that do not necessarily recognize its potential cross-sectional and time-series variation. However, the magnitude and variability of defaults, especially during the global financial crisis, have emphasized the importance of obtaining more precise estimates of recovery rates, and explaining their variation across issues and issuers. It is now intuitively understood that recovery rates are potentially driven by many different factors: endogenous variables (such as the specific characteristics of the assets of the firm and industry), or exogenous factors (such as overall macroeconomic conditions and market liquidity). It is important, therefore, to document the determinants of this risk factor, and to analyze their interaction effects with other dimensions of default risk. We investigate these relationships at the issue and obligor levels, for the US corporate bond market.

Specifically, we provide a detailed analysis of the microstructure of trading in defaulted bonds, working with a complete set of default events over the most recent decade, offering crucial insights. The study of market prices and trading behavior around different default events is important as many institutional investors are directly exposed to these post-default prices, e.g., because they have to immediately liquidate their positions, deliver the bonds through the settlement of credit default swaps (CDS) positions, or mark down the values of the defaulted bonds on their balance sheets. Furthermore, the examination of market prices provides us the opportunity to analyze all default events (including, e.g., distressed exchanges), and not only the outcomes of formal bankruptcy procedures, often revealed only years after the actual filing dates. Overall, this analysis allows us to discuss trading activity at different stages following default and to derive market-based estimates of recovery rates, which are of fundamental relevance to various market participants.

Trading activity reveals temporary sell-side pressure after default

We examine the trading activity of the defaulted bonds, as defined by traded prices and volumes, in a time window starting 90 days, before and ending 90 days after, the observed default event date (see Figures 1 and 2). We find that, although the price level is already rather low before the default event, the traded price falls significantly to its lowest level on the default day itself, to around 35% of face value, on average. The price recovers, in the first 30 days following default, to about 42% of face value and shows a less volatile evolution, thereafter. Furthermore, we find that the trading volume of a defaulted bond is relatively high on the default event day, providing evidence of temporary sell-side pressure, as prices are low on the default event date. This high level of trading activity dies down, within the first 30 days after default, to pre-default levels. Thus, this time window apparently represents the relevant trading period following default, in which investors split up and sell larger positions in defaulted bonds. Based on these findings, we define the recovery rate of a defaulted bond as the average daily traded price per unit of face value, over the default day and the following 30 days, covering the phase of high trading activity, as we conjecture that the price evolution in this time window is mostly driven by the default event itself.

1: Trading activity in defaulted bonds
This figure shows average trade prices on the default day and in the time window from 90 days before to 90 days after default.
2: Trading activity in defaulted bonds
This figure shows average volumes, as well as the average number of trades per bond, on the default day and in the time window from 90 days before to 90 days after default.

Default event type, industry and seniority drive recovery rate

Recovery rates are studied across bonds along various dimensions.

3: Chapter 11 restructuring and distressed exchange
This figure shows average trade prices for Chapter 11 restructuring on the default day and in the time window from 90 days before to 90 days after default.


First, we analyze them across different default event types, revealing that distressed exchanges have the highest recovery rates of 51.3% on average, whereas bankruptcy filings show significantly lower recoveries with an average of 37.1% (Figures 3 and  4 shows the evolution of prices for these two types of default events.) This finding provides evidence that bondholders are confronted with lower recoveries in formal legal procedures compared to out-of-court restructurings.

4: Chapter 11 restructuring and distressed exchange
This figure shows average trade prices for distressed exchanges on the default day and in the time window from 90 days before to 90 days after default.

Second, we find significant differences in recoveries between the default grades of the major rating agencies, which represent payment defaults and unlikely-to-pay events, respectively; in particular, the rating frameworks of Moody's and Fitch seem to incorporate recovery rate information to a greater extent than that of Standard and Poor's, which is focused on the probability of default. Third, we find that, among non-financial industries, utility and energy-related firms recover the most in default (e.g., electricity 48% or oil & gas 44.4%), while retailers recover the least at 33.4%. Interestingly, among financial firms, the banking and credit & financing industries recover the most in default (56.6%), whereas the financial services industry recovers the least (10.6%). Fourth, in terms of seniority levels we find, as expected, that secured bonds (49.3%) recover more than unsecured (39.1%) and subordinated bonds (15.1%). Fifth, we document substantial time-series variation in recoveries, as shown in Figure 5.

5: Recovery rates over time
This figure shows the time-series of average recovery rates (quarterly moving average) in the US corporate bond market over the period from July 2002 to October 2010.

Many other factors, especially liquidity, are important determinants

In the main part of our analysis, we employ regression models to explain the variation in recovery rates, using a comprehensive set of bond characteristics, balance sheet ratios, macroeconomic variables and liquidity measures. To begin with, we quantify the liquidity of defaulted bonds, applying different measures in our analysis, and explore their implications for recovery risk. These implications turn out to be of particular importance, since defaulted bonds are potentially illiquid. Consequently, we study to what extent changes in the underlying liquidity, following default, account for the observed post-default price evolution, as default might induce pressure on prices.

Overall, our regression analysis explains 66% of the total variation in recovery rates, with all four groups of variables contributing to the explanatory power. We demonstrate a clear link between the defined bond-specific liquidity measures and their recovery rates. The analysis reveals that trading in defaulted bonds is extremely costly. In particular, when measuring the transaction costs of trading using the price dispersion measure, estimated average transaction costs in defaulted bonds are 280 bp—about seven times compared to around 40 bp for non-defaulted bonds. Moreover, in our analysis, we document that illiquid bonds with higher transaction costs recover significantly less following default—an increase by 100 bp in transaction costs leads to a decrease in recovery rates by around 5% of face value.

Analyzing bond characteristics, we find that bonds that can be delivered into a CDS contract have a significantly higher recovery rate, possibly because of increased demand from protection buyers, who are required to physically deliver the underlying bond. In addition, we find that bond covenants significantly affect the level of the recovery rate. In particular, investment and financing covenants that provide protection for existing bondholders against potential adverse firm actions are important determinants. That is, restrictions on the investment and financing policy are effective tools that creditors can use to increase their recovery rates.

As for the other firm characteristics, among balance sheet ratios, we find significant effects for those ratios that are motivated by structural credit risk models, i.e., the higher is the equity ratio, and the lower the default barrier, the higher will be the recovery rate. Analyzing macroeconomic variables reveals a particularly strong effect for the market-wide and industry-specific default rates. Thus, a high default rate in the market as a whole, a systematic risk factor, or a high industry-specific default rate, as an indicator of industry distress, are both linked to significantly lower recovery rates for individual bonds, following default. Along the same lines, we find a positive relation between short-term interest rates, an indicator of the business cycle, and recovery rates.


In this paper, we examine the recovery rates of defaulted bonds in the US corporate bond market over the time period from 2002 to 2010. We provide a comprehensive analysis, going beyond the results that have been presented in the prior literature. We study the microstructure of trading activity and offer detailed insights into the stochastic nature and drivers of recovery rates by analyzing a broad set of explanatory variables, rather than only providing evidence on the effects of any one factor. We document temporary price pressure with high trading volumes on the default day and the following 30 days, and low trading activity at pre-default levels thereafter. Based on these observations, we quantify and analyze market-based recovery rates in the period representing the high trading activity window are estimated. We explore the relation between the recovery rates and these measures, considering additionally a comprehensive set of bond characteristics, firm fundamentals and macroeconomic variables. Our results on the effects of liquidity are particularly noteworthy, since they highlight the effects of liquidity on recovery rates.


Guillaume Plantin
Shadow banking and bank capital regulation
Review of Financial Studies | Volume 28, Issue 1 (Jan 2015), 146-175

The term “shadow banking system” refers to the institutions that do not hold a banking license, but perform the basic functions of banks by refinancing loans to the economy with the issuance of money-like liabilities. Roughly speaking, licensed banks refinance the loans that they hold on their balance sheets with deposits or interbank borrowing, whereas the shadow banking system refinances securities backed by loan portfolios with quasi-deposits such as money market funds shares.

Absent any prudential regulation, leverage—the fraction of total assets that is refinanced with such money-like liabilities—is much higher in the shadow banking system than in the licensed one. This higher leverage in turn makes the shadow banking system less stable. It is now well understood that the U.S. shadow banking system was the epicentre of the global financial crisis that erupted in 2008. An initial “run” on shadow institutions then propagated throughout the entire global banking system.

Perhaps surprisingly, the main response to the 2008 banking crisis consists thus far of a global trend towards increasing capital requirements for licensed banks, leaving many aspects of shadow banking unaddressed by regulatory reforms. These heightened capital requirements for licensed banks may trigger even more regulatory arbitrage than was observed in the recent past, thereby inducing a large migration of banking activities towards the shadow banking system. The higher solvency of the licensed banking system may then be more than offset by such growth in shadow banking, ultimately increasing the aggregate exposure of the money-like liabilities issued by both the formal and shadow banking sectors to shocks on loans.

The model

In “Shadow banking and bank capital regulation,” I offer a model of optimal banking regulation in the presence of regulatory arbitrage that can be used to assess this concern. The paper offers two mo delling contributions:

  1. It first builds a theory of bank capital regulation on the simple premise that the shareholders of banks, unlike those of other firms, internalize only a fraction of the total costs induced by a default on their liabilities. This in turn stems from the specific role of bank liabilities as money. Default by a given bank affects not only the depositors of this bank, but also other agents in the economy willing to trade with them using deposits as media of exchange. This implies that the optimal leverage of banks is lower than the one that bank shareholders would find privately optimal. In short, it is the difference in the nature of their liabilities that explains why the capital of banks is regulated while that of non-financial firms is not.
  2. The paper then introduces regulatory arbitrage by simply assuming that the regulator cannot monitor all the markets in which bank assets are refinanced with money-like liabilities. The small, but important, difference between the regulated and unregulated markets is that banks cannot commit not to use their material information about their assets in the latter because they are more opaque.
This modelling of shadow banking as unregulated banking captures parsimoniously the essence of regulatory arbitrage by banks. Regulatory arbitrage typically amounts to finding alternative legal and accounting classifications

for transactions that would be privately uneconomical given regulation under the standard classification, so that these transactions can be carried out outside the scope of regulation. For example, having money market funds with a fixed net asset value investing in commercial paper backed by asset-backed securities is economically close to the financing of loans by deposits, but legally, and therefore prudentially, quite different. In practice, these regulatory arbitrages exploit fine details and subtle loopholes in accounting rules and prudential regulations. These details vary over time, but the principle remains. Solving this model generates three main insights.

The insights

First, bank shareholders seek to bypass capital requirements and increase leverage on a loan portfolio in the shadow banking sector for two reasons. They may want to free up capital, or they may seek to exploit negative proprietary information about the portfolio. Heightened capital requirements for licensed banks make the former motive more likely than the latter, and thus alleviates informational frictions. As a result, tightening capital requirements spurs liquidity in the shadow banking sector.

Capital requirements should be lower with shadow banking

Second, the residual illiquidity premium in the shadow banking sector implies that a bank must transfer more risk per dollar raised in this sector than through its balance sheet. If capital requirements are tighter, banks substitute dollars raised through their balance sheets with dollars raised through the shadow banking sector. Because they transfer more risk per dollar raised by doing so, tightening capital requirements is overall counterproductive. It leads to an increase in the effective total leverage on loan portfolios. This implies that the optimal capital requirement for licensed banks is lower in the presence of a shadow banking system is typically lower than it would be if capital regulation was perfectly enforced.

Tightening Capital Requirements with shadow banking lowers the quality of collateral for unsecured creditors

Finally, if banks have more granular information than regulators about the riskiness of their assets, then tightening capital requirements in the presence of shadow banking also comes at the cost of an excessive encumbrance of their least risky assets. Banks find it preferable to refinance these safer assets in the shadow banking sector because their private information is less problematic for such assets that are therefore better collateral. Unsecured creditors, such as depositors (or deposit insurance funds), are then left with lower quality collateral.

This paper is an attempt at taking the possibility of imperfect enforcement seriously in a model of bank regulation. The motivation is that the banking industry devotes important resources to regulatory arbitrage, and that it is difficult for supervisory authorities to match these resources. This was particularly obvious in the years leading to the 2008 crisis. There is little evidence that enforcement and supervision have been much strengthened by financial reforms since then. Regulatory arbitrage is thus likely to remain an important dimension of banking, and realistic economic models of financial regulation should take this dimension into account.


Marti G. Subrahmanyam, Dragon Yongjun Tang, and Sarah Qian-Wang
Does the Tail Wag the Dog?: The Effect of Credit Default Swaps on Credit Risk
Review of Financial Studies | Volume 27, Issue 10 (Oct 2014), 2927-2960

The credit default swaps (CDS) market experienced exponential growth in 2000s, as it gradually became an effective tool for hedging, transferring and trading credit risk. However, the 2008 credit crisis highlighted the potential deleterious effects of over-the-counter derivatives on the real economy. It became abundantly clear that CDS may change the debtor-creditor relationship if creditors hedge or over-hedge their exposure by purchasing CDS of their borrowers. As a consequence, CDS-protected creditors have greater bargaining power and tend to be tougher in the process of restructuring. They may even have an incentive to push the borrowing firm into bankruptcy to trigger a payment from their CDS position, which increases the bankruptcy risk of the firm, ex-ante. we empirically examine the effects of CDS trading on the credit risk of the reference firms.

Credit risk increases after CDS introduction

We form our sample of CDS firms from multiple, leading data sources, including the GFI Group, CreditTrade, and Markit using transactions data, augmented with dealer quotes. We further construct the list of bankruptcies from New Generation Research's Public and Major Company Database (www.BankruptcyData.com), the Altman-NYU Salomon Center Bankruptcy List, the Mergent Fixed Income Securities Database (FISD), the UCLA-LoPucki Bankruptcy Research Database, and Moody's Annual Reports on Bankruptcy and Recovery. We then link the bankruptcy dataset with our CDS sample so as to identify the bankrupt firms that had CDS trading prior to their bankruptcy filings. In our final sample, we have 901 firms that have CDS traded on their debt during 1997-2009, of which 60 subsequently filed for bankruptcy protection. Additionally, we rely on CRSP, Compustat, FISD, and S&P to obtain firm accounting, financial, or credit rating information.

In Figure 1, we compare the distribution of credit ratings in the year preceding the introduction of CDS trading (year t-1) with the rating distribution two years after that (year t +2), for all firms with such contracts traded at some point in our sample. Figure 1 displays a discernible shift to lower credit ratings after the introduction of CDS trading. Whereas the proportion of BBB-rated firms is approximately the same, both before and after CDS trading begins, the proportion of AA-rated and A-rated firms clearly decreases. At the same time, the proportion of non-investment-grade and unrated firms increases.

1: Credit rating distributions before and after CDS introductions.


There is a discernible shift to lower credit ratings after the introduction of CDS trading.

We then model the probability of bankruptcy and investigate if CDS firms indeed have higher bankruptcy risk. However, to infer whether CDS trading causes a decrease in credit quality, we must consider the possibility that firms may be selected into CDS trading based on certain characteristics. Moreover, CDS trading may be initiated on firms for which market participants anticipate an increase in credit risk. To address these concerns, we consider the joint determination of bankruptcy filings and CDS trading. We uncover the true effect of CDS trading by using instrumental variables that have a direct effect on CDS trading, but only affect bankruptcy via their effect on CDS trading. We employ two such instruments. The first is the foreign exchange hedging positions of lenders and bond underwriters. Lenders with larger foreign exchange hedging positions are more likely to trade the CDS of their borrowers. The second is the lenders' Tier One capital ratio. Banks with lower capital ratios have a greater need to hedge the credit risk of their borrowers via CDS. We indeed find that both instrumental variables are significant determinants of CDS trading. It also appears valid to exclude both from the credit risk predictions of borrowing firms because they only affect borrowers' credit risk via CDS market activities. Our results indicate that CDS firms' credit risk indeed increases after the inception of CDS trading. The economic magnitude of the CDS effect on bankruptcy risk is also large: The marginal effect of CDS trading on the probability of a downgrade is 0.39%, whereas the average downgrading probability is 0.59%. The likelihood of bankruptcy for our sample firms in any year is 0.14%. However, for an average firm in our sample, the bankruptcy risk increases by 0.33% after CDS trading is initiated.

Default risk conditional on distress is higher for CDS firms

Creditors will become tougher in the course of renegotiations with financially distressed borrowers if their potential losses are protected by CDS. To provide further evidence for this channel, we test if firms in financial distress are more likely to face bankruptcy when they are referenced in CDS trading. We identify distressed firms based on their stock market performance. If a firm's stock return is in the bottom 5% of the market for two consecutive years, we classify it as financially distressed in the third year. We then compare the probability of bankruptcy filing between CDS firms and non-CDS firms in this distressed sample. The results indicate that once a firm is in financial distress, it is more likely to file for bankruptcy if it is referenced by CDS trading. This effect is economically large. Compared to the default rate of 17.71% for all financially distressed firms, the marginal effect of CDS trading is 19.84%.

More severe effect for CDS that exclude restructuring as credit event

An alternative test for the tougher creditor channel relates to the definition of credit events that trigger payment of CDS contracts. CDS protected creditors will clearly prefer firms to declare bankruptcy rather than restructure their debt only if their total payoffs are greater under bankruptcy than under restructuring. One such situation arises when bankruptcy, but not restructuring, triggers a credit event for CDS contracts and generates payments to the CDS buyers. Under this setting, creditors' incentives to encourage borrower bankruptcy will be even stronger if restructuring is not covered by the CDS contracts. CDS-protected creditors will not have this incentive to the same degree if their CDS contracts also cover restructuring as a credit event. As expected, we find the probability of bankruptcy is positively associated with a “no-restructuring” clause in a CDS contract, which provides additional evidence for the presence of tougher CDS-protected creditors.

Number of lenders increases after CDS introduction

In addition to tough creditors causing bankruptcy on an individual basis, creditor coordination is another important consideration when it comes to debt workout. If firms borrow money from a larger group of lenders after the inception of CDS trading, then creditor coordination will be more difficult, and bankruptcy more likely, when firms seek to restructure. Lead banks will likely not wish to appear to be driving their borrowers into bankruptcy, as the resultant long-run reputational damage may outweigh the short-run gains from their CDS position. However, other lenders, such as hedge funds or private equity players, who are not similarly constrained, may exploit CDS trading more intensively. Therefore, CDS trading may affect the size and composition of a firm's lenders. We investigate the impact of the introduction of CDS on the creditor relationships of a firm based on DealScan LPC data. We find that CDS trading significantly increases the number of lenders for a firm. CDS firms have 2.4 more lenders than non-CDS firms two years following CDS introduction. Moreover, consistent with the finding in previous literature, we document that a firm's bankruptcy risk increases with the number of banking relationships.


We provide evidence that the trading of credit default swaps increases reference firms' bankruptcy risk by creating tougher creditors and more diverse lender base. Our study reveals a real consequence of CDS trading and contributes to a better understanding of this important derivative market. Although our findings indicate that firms become more vulnerable to bankruptcy, once CDS are traded on them, we emphasize that this finding does not imply that CDS trading necessarily reduces social welfare. The benefits of the higher firm leverage that results from the mitigation of risk may, for some agents, outweigh the greater bankruptcy costs. Future work can examine the trade-off between the potential benefits and the bankruptcy vulnerability caused by CDS, shedding light on the overall impact of CDS trading on allocative efficiency.

Jayanthi Sunder, Shyam V. Sunder, and Wan Wongsunwai
Debtholder Responses to Shareholder Activism: Evidence from Hedge Fund Interventions
Review of Financial Studies | Volume 27, Issue 11 (Nov 2014), 3318-3342

With the separation of ownership and management in the typical modern corporation, shareholder activism provides a mechanism for dissatisfied owners to voice their concerns with the current management and attempt to make changes in the company. With dispersed ownership, activism by individual shareholders is rarely effective. An exception to this is when the activists are financially savvy hedge funds.

Increasingly, activist hedge funds have been thrown into the limelight following high profile interventions in household name companies. Other than management and shareholders, another important stakeholder in these corporations are lenders. When powerful shareholders such as hedge funds seek to make changes in the way a company is run, a natural question is how and to what extent debtholders are affected, since they constitute a stakeholder group with more limited control rights than either management or equityholders.

How do lenders react, on average, to hedge fund interventions?

In a related study, Klein and Zur (RFS 2011) document a negative reaction in prices of existing bonds when filings are first reported to the SEC by activist hedge funds. The average negative abnormal bond return of –3.9% stands in sharp contrast to the significant positive abnormal stock return, with estimates ranging from 7% to 10%.

Whereas existing providers of debt capital are relatively powerless in the face of activist shareholder interventions, new lenders have the ability to tailor new loan contracts which take into account the borrower's new circumstances. Banks, in particular, are sophisticated lenders who can infer activist intentions, and they can use price and non-price contract features to manage risk in lending. Using a sample of hedge fund interventions between 1995 and 2009 and the bank loans obtained by targeted companies in periods before and after intervention, we find that, on average, banks increase the interest rates of new loans taken post intervention, by 30 basis points (bps). This indicates that shareholder activism by hedge funds is detrimental to debt holders. However, the average effect disguises significant variation in the underlying changes in loan spreads. We also find that the type and mix of covenants in loan contracts become more stringent, consistent with our findings for loan spreads.

Do lenders ever benefit?

Greenwood and Schor (JFE 2009) document several objectives pursued by activist hedge funds, which can have different impacts on debtholders. In particular, one class of actions (which we categorize broadly as being about improving governance in the target firm) ought to benefit both shareholders and debtholders. We focus on two such actions. The first are outright attempts to replace an entrenched, valuedestroying CEO, in which case we observe decreases in loan spreads, amounting to 27 bps on average. The second type of governance-related action which is often encountered in hedge fund interventions are attempts to block merger transactions. In these types of interventions, the average loan spread decreases by 32 bps. For these two classes of interventions, it appears that shareholder intervention is viewed positively by debtholders.

Under what circumstances might lenders be worse off?

In addition to blocking mergers, activist hedge funds also intervene to attempt to force the firm into being acquired. Greenwood and Schor find that such interventions account for most of the positive abnormal stock returns which occur on announcement of activism by hedge funds (made public through mandated filings with the SEC). Several prior studies suggest that takeovers are usually viewed negatively by lenders to the acquired firms because of increased financial risk in targeted firms (for example, Cremers, Nair, and Wei (RFS 2007) and Chava, Livdan, and Purnanamdam (RFS 2009)). As expected, we find a very significant increase in loan spreads in these firms, to the tune of 78 bps. Another frequent demand by shareholder activists is for increased payouts, in the form of special dividends or share repurchases. All else equal, such changes in capital structure resulting in reduced stakes by equityholders is clearly not in debtholders' interest. We find an average increase in loan spreads of 34 bps for this type of intervention, reflecting the negative effect of such demands on the target firms' creditors.

What was different during the financial crisis years (2007-2009)?

Our sample covers activist interventions from 1995 to 2009. Since we are comparing loans before and after interventions, a large fraction of loans falling in the financial crisis period (2007-2009) would be classified as post-intervention. Any characteristics of loans specific to the financial crisis period which systematically differ from non-financial crisis periods could therefore affect interpretation of our results. In order to address these concerns, we conduct separate analyses of the two time periods, and the crisis period analysis shows that lenders' responded more negatively to takeover and financial restructuring actions, while the positive reactions to managerial entrenchment related actions were somewhat muted but still consistent with the overall sample. These findings are broadly consistent with the shock to financial markets resulting in greater conservatism on the part of lenders during the crisis period.

Does the strength of pre-existing lender protection matter?

We consider several measures of pre-existing takeover risk and managerial entrenchment in the target firms to examine whether they are correlated with loan spread changes. In line with our expectations, in firms pushed by the activist hedge funds into being acquired, the increase in loan spreads is higher when the firm was less vulnerable to being taken over prior to intervention. Similarly, in cases where the hedge funds' objective is to reduce managerial entrenchment, the decrease in spreads in greater for firms with ex- ante greater entrenchment.

We conduct several additional analyses in which we document a coinsurance effect, the data suggesting that takeovers involving acquirers with better credit quality than the target are associated with a reduction in loan spreads, a result which goes in the opposite direction to the overall increase in loan spreads which we observe for cases involving the activist-targeted firm being pushed into being acquired.


In conclusion, this paper contributes to a better understanding of the consequences of hedge fund shareholder activism as viewed from debtholders' perspective. We show that different objectives pursued by activists have different implications for debtholders, and we specify different sets of circumstances in which the effects are either beneficial or adverse. Thus, shareholder activism does not always exacerbate bondholder-shareholder conflicts of interest. In particular, when shareholder activists address shareholdermanager agency problems and reduce the credit risk of the target firm, debtholders view the actions favorably.


Nishant Dass, Omesh Kini, Vikram Nanda, Bunyamin Onal, and Jonathan Wang
Board expertise: Do directors from related industries help bridge the information gap?
The Review of Financial Studies, Volume 27, Issue 5 (May 2014), 1533-1592

Boards of directors play a central role in corporate control and decision-making, attracting both media and regulatory attention. Less evident perhaps is the advisory function of boards, where the industry expertise and connections of individual directors can enhance the quality of strategic advice provided to management. Firms may choose directors who can help alleviate uncertainty in the economic environment and lower transactions costs in dealing with external entities. We argue that directors drawn from a firm's supply-chain industries, with their related industry knowledge and expertise, can be particularly valuable.

In a large sample of publicly-listed firms, we find that about 60% of firms have had a director who is also an officer or a director of another firm in their supply chain industries; we denote these directors as “directors from related industries” (“DRIs”). These directors can bring potentially valuable knowledge about their own industries as well as facilitate the firm's access to contacts in those industries. This knowledge helps the firm overcome information challenges such as anticipating industry conditions and trends, thereby facilitating better management of its factors of production and protecting it against demand or supply shocks. By providing information about supply chain industries, DRIs can also improve the board's ability to monitor by narrowing the information gap between the board and the firm's management. We build on these arguments by investigating the following questions:

Why do some firms choose to have DRIs?

To answer this question, we propose and empirically validate the following hypotheses: Information-related hypothesis. Firms that face a significant information challenge with respect to their supply chain will benefit more from the industry expertise, knowledge, and networks of DRIs. For instance, innovative firms may favor DRIs since they require specialized inputs and the demand for their output is harder to predict. In addition, the lack of other sources of information and the strength of economic links between the firm's industry and its supply chain industries will increase the demand for DRIs. We find that the prevalence of DRIs at firms is consistent with this hypothesis. For instance, Table 1 reports the top-5 and bottom-5 industries, when ranked by the percentage of firms with at least one DRI on the board. It suggests that firms operating in industries that face the largest (smallest) information challenges are the most (least) likely to utilize the services of DRIs.

1: Industries ranked by the prevalence of DRIs (over the sample period 1990-2005)
Top 5 Number of firms in Percentage of firms
SIC description the given industry with at least one DRI
Biological not diagnostic 1,063 78.65%
Computer storage devices 319 62.07%
Advertising agencies 108 73.15%
Electrical instruments 496 61.09%
In vitro/vivo diagnostics 656 66.16%
Bottom 5 Number of firms in Percentage of firms
SIC description the given industry with at least one DRI
Newspapers 283 17.67%
Savings 770 10.78%
Poultry 112 5.36%
Home health care 209 8.61%
Electric housewares 104 10.58%

Further, consistent with the information-related hypothesis, we find that characteristics such as R&D intensity, patent grants, patent citations, and the differentiated nature of an industry's products increase the likelihood, while characteristics such as stock price informativeness reduce the likelihood of having DRIs on the board.

Market structure hypothesis.
It will be easier for firms with greater market share to attract DRIs because these firms are likely to be relatively more important as sources of information and network connections for the firms from which the DRIs are drawn. Further, common ownership to reduce coordination and/or contracting costs with potential/actual suppliers and customers is more likely to receive antitrust scrutiny in concentrated industries. Hence, firms in concentrated industries may, at least partially, achieve the same objective by appointing DRIs. Consistent with the market structure hypothesis, we find that the likelihood that a firm has DRIs is increasing in its market share and industry concentration. We also find that the presence of DRIs is greater in industries with higher degrees of vertical integration. This result suggests that DRIs can alleviate the high coordination costs and contractual frictions that are likely present in these industries.
Conflicts of interest hypothesis.
Having DRIs on the board also has a potential downside arising from conflicts of interest. For instance, DRIs from current or potential customers/suppliers may influence the firm's management to improve the prices/terms negotiated or garner new business for their affiliated firms. Our evidence is consistent with this hypothesis. Specifically, it is uncommon for DRIs to come from the current customers/suppliers of the firm. Further, it is rare for the DRIs' affiliated firms to eventually become customers/suppliers of the firm. Finally, the likelihood of DRIs being present is higher when the firm is less vulnerable to DRIs' attempts to promote their affiliated firms.

Does the presence of DRIs contribute to firm value/performance?

We next turn to the second question by examining the impact of DRIs on firm value and performance. We find that the presence of DRIs appears to improve firm value and performance. For instance, the presence of a DRI is associated with a 5.5% higher ROA. The stock market reaction upon the appointment of a DRI to the board is also positive—e.g., the risk-adjusted abnormal return over a three-day window centered around the announcement of a DRIappointment is 2.5%.

Clearly, not every firm will benefit from having DRIs. Consistent with the informational benefits of having DRIs, we find that their effect on the firm's value is stronger for firms operating in more innovative environments and those with less informative stock prices. Further, the impact of DRIs on firm value is stronger for firms with greater market share, which is consistent with the notion that these firms are less influenced by the DRIs' conflicts of interest and that the presence of DRIs can enable dominant firms to foreclose their rivals from critical inputs/outlets. Finally, the value impact of DRIs is only significant when the CEO is also the Chairman, which is consistent with powerful CEOs being less affected by the conflicts of interest that arise due to the DRIs trying to benefit their affiliated firms.

What are the channels through which DRIs benefit the firm?

We address the third question by examining some specific channels through which DRIs can add value to the firm. We find that the information/expertise of DRIs helps firms: (i) respond better to industry sales shocks, (ii) alleviate financial constraints, and (iii) shorten their cash conversion cycle primarily through better management of inventories. For example, the presence of a DRI is associated with a cash conversion cycle that is shorter by at least 21 days and an inventory conversion period that is shorter by 18 days.

On a cautionary note, not every firm can hope to benefit by having DRIs on its board. There are several costs associated with having DRIs and these may outweigh their potential benefits. First, when DRIs are from actual or potential suppliers and customers, there is reason to be concerned about conflicts of interest. Second, DRIs can be a potential source of proprietary information leakage to rival firms. Finally, bringing in a DRI comes at the cost of excluding another potential director, who can add value along other dimensions (e.g., by providing political connections or financial expertise).

Not an AFA President.... ...but the editor of this jagazine. We ran out of volunteers, and felt it was just fair to expose ourselves, too.

Not an AFA President.... ...but the publisher of this jagazine. Ivo's picture was funnier than Bhagwan's.