Some Recent Trends in
Empirical Corporate Finance Research
Sheng-Syan Chen
Department of Finance
College of Management
National Taiwan University
(1)
Managerial Overconfidence
• Overconfident managers tend to think they are better than they
really are in terms of ability, judgment, or prospects for successful
life outcomes (“optimism”).
Overconfident managers are likely to overestimate the net
discounted expected payoffs from uncertain endeavors, either
because of a general tendency to expect good outcomes, or
because they overestimate their own efficacy in bringing about
success.

Overconfident managers tend to overestimate the precision of their
own forecasts or underestimate the volatility of random events,
resulting in subjective probability distributions that are too narrow.
(2)
 Empirical evidence
(3)
 Are overconfident CEOs better innovators?
(Hirshleifer, Low & Teoh (HLT), 2012, JF)
• People tend to be more overconfident about their performance on
hard rather than easy tasks (Griffin & Tversky 1992).
Thus, relatively overconfident CEOs are expected to be especially
enthusiastic about risky, challenging, and talent- and visionsensitive enterprises.
Innovative projects—which apply new business methods, develop
new technologies, or offer new products or services—are risky
and challenging.
Firms with overconfident managers are expected to accept greater
risk, invest more heavily in innovative projects, and achieve
greater innovation.
(4)
• Measuring CEO overconfidence

Options-based measure:
It is typically optimal for risk-averse, undiversified executives to
exercise their own-firm stock options early if the option is
sufficiently in the money (Hall & Murphy 2002).
Confident CEO (Options) takes a value one if a CEO postpones
the exercise of vested options that are at least 67% in the money,
and zero otherwise (Malmendier & Tate (2005a, 2008)).
As there are no detailed data on a CEO’s options holdings and
exercise prices for each option grant, HLT calculate the average
moneyness of the CEO’s option portfolio for each year (Campbell
et al. (2011)).
(5)
First, for each CEO-year, HLT calculate the average realizable
value per option by dividing the total realizable value of the options
by the number of options held by the CEO.
The average exercise price of the options is calculated as the fiscal
year-end stock price minus the average realizable value per option.
The average percent moneyness of the options equals the per-option
realizable value divided by the estimated average exercise price.
As they are only interested in options that the CEO can exercise,
HLT include only the vested options held by the CEO.
(6)
6

Press-based measure:
Following Malmendier & Tate (2005b, 2008) and Hribar & Yang (2011), HLT
search for articles referring to the CEO.
For each CEO and year, HLT record:
(1) total number of articles;
(2) number of articles containing the words confident, confidence, or variants
(overconfidence and over-confident);
(3) number of articles containing the words optimistic, optimism, or variants
(overoptimistic and over-optimism);
(4) number of articles using pessimistic, pessimism, or variants (over-pessimistic);
(5) number of articles using reliable, steady, practical, conservative, frugal,
cautious, or gloomy.
For each year, HLT compare the number of articles that use the “Confident”
terms (categories 2 & 3) and the number of articles that use the “Cautious” terms
(categories 4 & 5).
(7)
HLT measure CEO overconfidence for each CEO i in year t as
where ais is the number of articles using the Confident terms and bis
is the number of articles using the Cautious terms.
Following Malmendier & Tate (2008), HLT also control for the
total number of press mentions over the same period
(TotalMention).
The press may be biased toward positive stories and this would
imply a higher number of mentions as confident or optimistic when
there is more attention in the press.
(8)

Measuring innovation
HLT measure resource input into innovation with R&D scaled by
book assets.
Their output-oriented measures of innovation are based on patent
counts and patent citations.
Owing to the finite length of the sample, citations suffer from a
time truncation bias. HLT follow Hall et al. (2001, 2005) and
adjust the citation count of each patent in 2 different ways:
Qcitation count: Each patent’s citation count is multiplied by the
weighting index from Hall et al. (2001, 2005).
TTcitation count: Each patent’s citation count is scaled by the
average citation count of all patents in the same technology class
and year.
(9)

Sample

Since CEO overconfidence measures are lagged by one year, HLT
require that the CEO be the same one in the prior year to ensure
that the characteristics of the CEO in place are observed at the time
the innovation is being measured.
Financial firms and utilities are excluded.
The final sample consists of 2,577 CEOs from 9,807 firm-year
observations between 1993 and 2003.
Of these observations, 8,939 firm-years have information on the
options-based measure, while 7,762 firm-years have information on
the press-based measure of overconfidence.
(10)
Steve Jobs of Apple Computers turns out to be overconfident using
both measures of overconfidence.
(11)
(12)
(13)
(14)
(15)
Note: An industry is defined as innovative if the average Qcitation count per patent for the industry is
greater than the sample median.
(16)
(17)
• The greater innovative output for given R&D input achieved by
overconfident CEOs does not necessarily translate into higher firm
value.
Hall et al. (2005) show that patent citations are positively
correlated with firm value, but overconfident CEOs could be
overpaying to achieve increased citation counts (possibly using
resources other than R&D expenditures), reducing firm value.
A possible way to address this issue is to regress firm value on
CEO overconfidence or the innovation that results from it.
However, such a test is subject to endogeneity problems.
HLT hence examine a more specific issue, whether overconfidence
allows firms to translate growth opportunities into realized firm
value (measured by Tobin’s Q).
(18)
• Following Bekaert et al. (2007), HLT use the industry price to
earnings (PE) ratio as an exogenous instrument for firm growth.
HLT calculate the monthly industry PE ratio as the logarithmic
transformation of the ratio of the industry’s total market
capitalization to the industry’s total earnings.
PE is affected by risk as well as growth opportunities, so their tests
are weakened by noise to the extent that the PE ratio is influenced
by discount rate changes.
As in Bekaert et al. (2007), HLT hence subtract the 60-month
moving average of the PE ratio, motivated by the idea that
discount rates are more persistent than growth opportunities.
Finally, HLT average the difference over the fiscal year to form
our measure of exogenous growth opportunities.
(19)
Overconfident CEOs are more effective at exploiting growth opportunities
and translating them into firm value, especially among innovative industries.
(20)
Supply Chain
• Wealth effects of corporate decisions on announcers:
 Short-term market reactions
 Long-run performance
• Wealth effects of corporate decisions on industry rivals:







M&As: Eckbo (1983, JFE), Stillman (1983, JFE), Song & Walkling (2000, JFE),
Cai, Song & Walkling (2011, RFS)
IPOs: Hsu, Reed & Rocholl (2010, JF)
SEOs: Szewczyk (1992, JF), Slovin, Sushka & Polonchek (1992, JFE)
Share repurchases: Hertzel (1991, JF)
Bankruptcy: Lang & Stulz (1992, JFE), Benmelech & Bergman (2011, JF)
Going private: Slovin, Sushka & Bendeck (1991, JF)
Privatization: Eckel, Eckel & Singal (1997, JFE)
(21)
• Linkage between corporate decisions and suppliers and customers:
(22)
• Inter-firm linkages and the wealth effects of financial distress
along the supply chain:
(Hertzel, Li, Officer & Rodgers, 2008, JFE)
 Investigate the wealth effects of distress and bankruptcy for
filing firms’ suppliers and customers.
Lang and Stulz (1992)
(23)
• Identifying suppliers and customers
Hertzel et al. use the approach of Fee & Thomas (2004) by relying
on various financial accounting standards, which require firms to
disclose the identity of any customer representing more that 10%
of the firm’s total sales.
Because the bankruptcy filing firms account for more than 10% of
each supplier’s total sales, this procedure yields a set of suppliers
that is reliant on the filing firm.
Hertzel et al. use the same procedure to form the sample of filingfirm customers, but these customers may not be reliant on the
filing firms. Thus, they also examine a subset of reliant customers
defined as those for which purchases from the filing firm scaled
by total cost of goods sold is greater than 1%.
(24)
•
Sample
A preliminary sample of 1,695 bankruptcy filings between 1978
and 2004 is obtained from Bankruptcy DataSource Index.
The final sample includes a total of 250 filing firms that have at
least one customer or one supplier.
For the full sample of 250 filing firms, a total of 311 customers
and 275 suppliers are identified.
The analysis of supply-chain contagion effects is based on
separate portfolios of customers and suppliers for each bankruptcy
filing and for each pre-filing distress event.
The average numbers of customers and suppliers per bankruptcy
are 2.0 and 2.2 .
(25)
(26)
Identifying pre-filing distress
Compute abnormal return by using the
The average distress-day abnormal percent
The average filing-period abnormal
(Brown
and
returnMarket-adjusted
(dollar abnormal return)return
for filingmodel
firms
percent return
forWarner(1985))
bankrupt firms is
is 26% ($118M).
19%.
Distress Period
Filing Period
-5
-5
+5
Pre-filing Distress date
+5
Bankruptcy Filing Date
search the CRSP database over
the year prior to the Chapter 11
filing date and find the day on
which the filing firm has the most
negative dollar abnormal return.
(27)
(28)

Hertzel et al. also use the Fama & French (1993) three-factor
model and the Carhart (1997) four-factor model to examine
post-filing returns to suppliers and customers, and find no
evidence of investor underreaction.
(29)
(30)

Cross-sectional determinants of the returns to suppliers & customers


Empirical predictions:
Supplier & customer leverage:
To the extent that leverage reduces the flexibility of suppliers &
suppliers to respond to the distress of the filing firm, contagion
effects are expected to be larger for suppliers or customers with
greater leverage.

Product specificity:
Suppliers that sell specialized products to the filing firm (or
customers that purchase specialized products from the filing firm)
are more likely to experience contagion because of the likelihood
of greater contractual ties to the filing firm and higher costs of
rerouting outputs (or inputs). Use R&D intensity as a proxy for
the specialization of a firm’s product.
(31)

Filing firm industry concentration:
Suppliers & customers of filing firms in concentrated industries
are expected to have fewer switching alternatives for re-routing
supply or demand, respectively, and thus are more likely to suffer
greater contagion effects.

Outcome of the bankruptcy process:
The contagion effects to be greater when the filing firm does not
successfully emerge from the bankruptcy process as an
independent entity.

Empirical results:
None of multivariate regressions delivers statistically significant
results as anticipated above.
This may be due to the lack of statistical significance of tests to
small sample sizes, the considerable cross-sectional variation in
supplier & customer returns, and the coarseness of the proxies
used to capture the characteristics.
(32)