MANAGERIAL ABILITY AND

EMPLOYEE PRODUCTIVITY*
Dipankar Ghosh, Xuerong (Sharon) Huang
and Li Sun

ABSTRACT
Purpose – This study examines how managerial ability relates to employee
productivity using a broad and generalized sample of US firms.
Methodology – This study employs a generalized sample of firm-years from
all industries between 1980 and 2013.
Findings – By contending that managers differ in their ability to synchronize
management processes and human capital in ways that enhance employee pro-
ductivity, the authors provide evidence showing that more-able managers are
associated with higher employee productivity. In addition, the authors find that
high-ability managers moderate the negative relation between uncertain envi-
ronments (high-technology firms) and employee productivity. Furthermore, the
authors decompose employee productivity into employee efficiency components
and employee cost components. The authors find a significant positive associa-
tion between managerial ability and the employee efficiency component, but do
not see a significant association between managerial ability and the employee
cost component.
Value – The results contribute to the understanding of employee produc-
tivity by showing the relation between managerial ability and employee
productivity.
Keywords: Managerial ability; employee productivity; employee efficiency;
employee cost; financial distress; environmental uncertainty

Data are available from sources identified in this article.

*

Advances in Management Accounting, Volume 32, 151–180

Copyright © 2020 by Emerald Publishing Limited
All rights of reproduction in any form reserved
ISSN: 1474-7871/doi:10.1108/S1474-787120200000032006
151
152 DIPANKAR GHOSH ET AL.

1. INTRODUCTION
The supply of managerial services, particularly in the upper echelons of a firm,
has long been recognized as a critical requirement for firms to grow and diversify
(Helfat & Peteraf, 2015; Holcomb, Holmes, & Connelly, 2009; Penrose, 1959).
Research also recognizes that managers are idiosyncratic in their abilities to build,
integrate, reconfigure, and competitively reposition firms’ resources and capabili-
ties (Adner & Helfat, 2003; Butler & Ghosh, 2015; Dyreng, Hanlon, & Maydew,
2010; Kahneman & Tversky, 1984). Helfat and Peteraf (2015) attribute this ability
to “managerial cognitive capability” or the capacity to perform both physical and
mental activities. Despite the growing literature recognizing the idiosyncrasy of
managers and the normative connection between managers’ ability and employee
productivity, empirical studies in this area remain sparse.1
Initial insights on this relation from Penrose (1959, p. 5) observe that
the resources with which a particular firm is accustomed to working will shape the productive
services its management is capable of rendering … but also the experience of management will
affect the productive services that all its other resources are capable of rendering.

Barney (1991, p. 117) asserts that a manager’s ability to understand and

efficiently use the firm’s resources “has the potential for generating sustained
competitive advantages” for a firm. Managing human resources is centrally
important to the execution of firm-level strategy and improving employee pro-
ductivity (Koch & McGrath, 1996). However, there is no research concerning
managers’ operating abilities that forges a link between human capital resources
and their productivity.
Our core premise is that more-able managers can manage their employees more
efficiently than less able managers. Castanias and Helfat (2001, p. 665) argue that
“the skills of top management combined with other firm assets and capabilities
jointly have the potential to generate rent.” This statement suggests that possess-
ing valuable, rare, inimitable, and non-substitutable resources is a necessary but
insufficient condition for creating performance value. Indeed, the value is created
when resources are evaluated, manipulated, and deployed appropriately only by
managers within the firm’s context (Lippman & Rumelt, 2003). In addition, our
research explores when a manager’s ability matters most to employee productivity,
which is for firms operating in the high-technology industry or under uncertain
environments. In these circumstances, more-able managers can better recognize
changes in the business environment, resulting in decisions that are positively
reflected in the firm’s financial performance and enhance the value of the firm’s
resources.
In our current research, we initially hypothesize a positive relation between
managerial ability and employee productivity. Next, to further explore this rela-
tion, we follow the DuPont technique and decompose the employee productivity
ratio into the employee efficiency component and the employee cost component.
The employee efficiency component captures the income generated per unit of
employee cost, while the employee cost component measures the average com-
pensation cost per employee. We posit a positive relation between managerial
Managerial Ability and Employee Productivity 153

ability and the employee efficiency component and a negative relation between
managerial ability and the employee cost component.
To quantify managerial ability, we use the measure developed by P. Demerjian,
Lev, and McVay (2012),2 which distinguishes the effect of the manager from the
effect of the firm and is available for a large sample of firms. Also, we adopt
two other proxies for CEO ability used in the recent literature in our robustness
check. Our first alternative proxy for managerial ability is based on the number
of press citations for a CEO over a prior five-year period (Baik, Farber, & Lee,
2009; Rajgopal, Shevlin, & Zamora, 2006).3 Our second measure of CEO ability
follows Rajgopal et al. (2006) and uses the three-year average of the cumulative
distribution function (CDF) of return on assets (ROA) for each CEO-firm-year
by industry. To quantify employee productivity, we follow Stuebs and Sun (2010)
and define employee productivity as the income generated per employee. In our
robustness check, we also adopt an alternative measure of employee productivity
(EMPPRO_ALT), which is the employee performance ratings provided by the
MSCI’s environmental, social, and governance (ESG) database.
Our study employs a generalized sample of firm-years from all industries
between 1980 and 2013. In the first set of tests, we examine the relation between
managerial ability and employee productivity. We find a significant positive rela-
tion between managerial ability and employee productivity. This finding suggests
that more-able managers help enhance employee productivity. Also, we test this
relation using two alternative measures of managerial ability and one alternative
measure of employee productivity and still find consistent results supporting our
main hypothesis. We further investigate whether the managerial ability is associ-
ated with the two components of employee productivity, which are the employee
efficiency component and the employee cost component. We find a significant
positive relation between managerial ability and the employee efficiency com-
ponent, suggesting that more-able managers help improve employee efficiency
level. However, we do not find a significant relation between managerial ability
and the employee cost component. In the second set of tests, we explore the
circumstances when managerial ability matters more to employee productivity.
Moreover, we find that managerial ability has more influence on employee pro-
ductivity when the firm is in the technology industry, or the firm is under uncer-
tain environment.
This study makes several significant and noticeable contributions. First, our
core premise is valid; that is, managers differ in their ability to enhance employee
productivity. P. Demerjian et al. (2012) suggest that more-able managers under-
stand the technology and industry trends better; therefore, these managers are
better at manage firms in an efficient way. Our article extends this line of research
and examines explicitly one aspect of managerial ability, which is the ability to
manage employees. In addition, our study focuses on one specific firm perfor-
mance measure that is employee productivity. Second, our study shows the cir-
cumstances when and how managerial ability matters more. We show that when
the firm is in the technology industry or is under an uncertain environment, the
managerial ability has more impact on employee performance. Third, this study
154 DIPANKAR GHOSH ET AL.

adds to the debate on whether managers matter in firms’ decisions and outcomes.
Upper echelons theory (D. Hambrick, 2007; D. C. Hambrick & Mason, 1984;
Tushman & Romanelli, 1985) states that organizational outcomes are partially
influenced by managers’ differing background characteristics. Many studies in
accounting (e.g., Aier, Comprix, Gunlock, & Lee, 2005; Bamber & Wang, 2010;
Bertrand & Schoar, 2003; Francis, Nanda, & Olsson, 2008; Ge, Matsumoto, &
Zhang, 2011) document that individual managers matter in firms’ decisions and
outcomes. Our study contributes to this stream of research by showing that man-
agerial ability matters in employee performance. Fourth, despite the theoretical
and normative connection made between managerial ability and employee pro-
ductivity, to the best of our knowledge, our study is the first one that performs
a direct empirical test on the link between managerial ability and employee pro-
ductivity.
The remainder of this article is organized as follows: Section 2 reviews related
studies and presents hypotheses development. Section 3 presents the research
design, including the measurement of the dependent and primary independent
variables and the empirical specification. Section 4 discusses the results, including
the sample selection and descriptive statistics, while Section 5 provides additional
tests. Section 6 concludes this article.

2. LITERATURE REVIEW AND HYPOTHESES

DEVELOPMENT
2.1. Managerial Ability
Managerial ability is the knowledge, skills, and experience residing with and uti-
lized by managers (Hitt, Bierman, Shimizu, & Kochhar, 2001).4 From a strategic
perspective, managerial ability derives from two primary sources: domain exper-
tise and resource expertise. Domain expertise refers to a manager’s understanding
of the industry context and the firm’s strategies, products, markets, task environ-
ments, and routines (Boeker, 1989; Kor, 2003; Spreitzer, McCall, & Mahoney,
1997). These skills differ across different firms and industry domains. The more
specific the ability embedded in managers, the more unlikely it is to be entirely
transferable to other firms and to challenge rivals to imitate, making it poten-
tially a source of superior performance (Hatch & Dyer, 2004). Resource expertise
manifests through experience with resource management processes (Sirmon et al.,
2007) to realize an organization’s competitive advantage.
In support of these views, researchers link managerial ability directly to many
performance outcomes. P. Demerjian, Lev, Lewis, and McVay (2013) examine
the relation between managerial ability and earnings quality. They find that
more-able managers are associated with fewer subsequent restatements, higher
earnings and accruals persistence, lower errors in the inadequate debt provi-
sion, and higher quality accrual estimations. Baik, David, and Lee (2011) find
a positive relation between CEO ability and management earnings forecast issu-
ance. Wang (2013) examines the informativeness of insider trades conditional
on managerial ability and finds more-able managers have higher net insider sales
Managerial Ability and Employee Productivity 155

before the earnings break than do less-able managers. P. R. Demerjian, Lewis-

Western, and McVay (2015) find that more-able managers are more likely to
intentionally smooth earnings, and they are more likely to do so when smooth-
ing benefits shareholders, the manager, or both. Krishnan and Wang (2015) show
negative relations between managerial ability and both audit fees and going
concern options, suggesting managerial ability plays an essential role in audi-
tors’ decisions. Libby and Tan (1994) argue for and find evidence of a relation
between the learning environment and the importance of the problem-solving
ability to auditors’ judgment performances. Other work links managerial ability
to outcomes such as profitability and growth (Carpenter, Sanders, & Gregersen,
2001; Geletkanycz & Hambrick, 1997; Hitt et al., 2001; Miller & Shamsie, 1999).
Firms consist of people whose abilities can vary widely, and this variation among
individuals matters far more in organizational performance than is assumed or
acknowledged (Mollick, 2012).

2.2. Employee Productivity

The resource management process is the ability of managers to select and con-
figure a firm’s resource portfolio, bundle resources into unique combinations,
and deploy them to exploit opportunities in specific contexts. For human capital
resources, constraints limiting effective use increase as proficiencies across differ-
ent skill sets and abilities embedded within individuals vary (Coff, 1997; Lado
& Wilson, 1994; Wright & Snell, 1998). Instead of maximizing an individual’s
skill set or ability, managers must consider their marginal contribution, relying on
knowledge of their skills and abilities to combine them in ways that enhance the
value-creating potential of the bundle (Alchian & Demsetz, 1972).
Employees are essential to the success of an organization (e.g., Berk, Stanton, &
Zechner, 2010; Carlin & Gervais, 2009; Rajan & Zingales, 1998). They are criti-
cal value drivers and highly related to organizational strategy (C. D. Ittner &
Larcker, 2001). Edmans (2011) documents empirical evidence to argue that com-
panies should focus on their employees as critical assets. Prior studies also suggest
that managers play a significant role in managing employee performance. For
example, Milner and Pinker (2001) find that managers can influence employee
contracting which affects employee productivity. Davis and Albright (2004) find
that bank managers’ use of a balanced scorecard increases employee perfor-
mance. Using Belgian firms, Dierynck, Landsman, and Renders (2012) document
that managers increase employee costs to a smaller extent for sales increases but
decrease employee costs to a more significant extent for sales decrease, in order
to meet or beat their zero earnings benchmark. Following this line of literature,
we are interested in analyzing whether managerial ability plays a role in employee
performance.

2.3. Hypotheses Development

Prior research on employee productivity documents that human resource man-
agement economically and statistically impacts employee productivity signifi-
cantly (Huselid, 1995). Consistent with this notion, Koch and McGrath (1996)
156 DIPANKAR GHOSH ET AL.

show that firms with more sophisticated human resource strategies positively and
significantly impact employee productivity. This relation is more pronounced in
large capital-intensive organizations compared to small capital-intensive organi-
zations. Furthermore, Ichniowski, Shaw, and Prennushi (1997) suggest that some
innovative human resource practices, such as incentive pay, teams, flexible job
assignments, employment security, and training achieve higher employee produc-
tivity than traditional approaches. Another line of research on employee pro-
ductivity focuses on employees’ effects. On the one hand, research shows that
the frequency of favor exchanges among peer employees has a positive impact
on employee productivity (Flynn, 2003). On the other hand, research shows that
employee participation improves employee productivity (Bhatti & Qureshi, 2007).
However, increasing employee participation is a long-term process that requires
both top managers’ attention and employees’ efforts.
In addition to human resource management and employees’ effects,
recent research shows that manager and employee relationships affect
employee productivity. For example, Mayer and Gavin (2005) find that
employees’ productivity improves with employees’ trust in their top man-
agement team. Bhatti and Qureshi (2007) also suggest that the shared goals
between managers and employees help improve employee productivity.
Although existing research suggests that managers play an essential role in
employee productivity, concentrating on the manager and employee rela-
tionship assumes managers are idiosyncratic and ignores any difference
in managers. Following recent research on managers’ style (Bertrand &
Schoar, 2003) – specifically managerial ability (P. Demerjian et al., 2012) –
we assume that managers have different ability levels and that their ability
levels influence how they view employee participation. More importantly,
managers’ ability levels influence how able they are to work with and earn the
trust of their employees, as well as how they strategize to meet their shared
goals. Therefore, we propose that managerial ability influences employee pro-
ductivity. This article extends prior literature on employee productivity by
incorporating managerial ability as another contributing factor in addition
to human resource management, employees’ effect, and the relation between
managers and employees. Based on the above discussion, we first hypothesize
the following:
H1. Managerial ability is positively related to employee productivity.
Using DuPont technique (e.g., Hopper & Armstrong, 1991; Little, Little, &
Coffee, 2009; Stuebs & Sun, 2010), we further decompose the employee produc-
tivity ratio into two components: the employee efficiency component and the
employee cost component. The employee efficiency component measures how
efficiently a firm uses total employee cost (XLR) to generate profits, while the
employee cost component measures how well a firm controls employee cost. If
more-able managers better understand and manage their employees, we expect
that these managers are better able to improve employee efficiency and reduce
employee cost. In other words, more-able managers can better utilize employee
cost to generate profits and identify better ways to reduce (compensation) cost
Managerial Ability and Employee Productivity 157

per employee. Thus, we posit a positive relation between managerial ability

and employee efficiency and a negative relation between managerial ability and
employee cost. We propose the following general hypotheses:
H2a. Managerial ability is positively related to the employee efficiency component.
H2b. Managerial ability is negatively related to the employee cost component.

3. RESEARCH DESIGN
3.1. Measurement of the Dependent Variable – Employee Productivity
and its Components
Consistent with Stuebs and Sun (2010), employee productivity is computed by
dividing net income (NI) before employee cost by the number of employees
(EMP). The formula is as follows:

Employee Productivity
(EMPPRO) =
( )
Net Income NI5 + Total Employee Cost(XLR6 )
Number of Employees (EMP 7 )

(1)

We follow the DuPont technique to decompose the employee productivity ratio

into two components:
Employee Productivity (EMPPRO) = Employee Efficiency (EMPEFF)
×Employee Cost (EMPCST)
(2)
where
Employee Efficiency
Net Income ( NI) + Total Employee Cost (XLR) (3)
(EMPEFF) =
Total Employee Cost (XLR)
Total Employee Cost (XLR) (4)
Employee Cost (EMPCST) =
Number of Employees (EMP)

Employee efficiency (EMPEFF) measures how efficiently a firm uses XLR to

generate profits, while employee cost (EMPCST) measures how well a firm con-
trols employee costs. Given the magnitude and possible skewness of the variables
EMPPRO, EMPEFF, and EMPCST, we use the natural log of EMPPRO (ln LP),
EMPEFF (ln LE), and EMPCST (ln LC) in our empirical analyses.

3.2. Measurement of the Primary Independent Variable – Managerial Ability

We use the managerial ability score (MASCORE) developed and validated by
P. Demerjian et al. (2012) as the main proxy for managerial ability in our study.
158 DIPANKAR GHOSH ET AL.

Their managerial ability measure is a performance-based measure of managers’

efficiencies in using their firms’ resources to generate revenue. P. Demerjian et al.
(2012) use a two-step approach to developing their managerial ability meas-
ure. First, they rely on data envelopment analysis (DEA) to estimate total firm
efficiency by industry and year. Given a collection of points in a multidimen-
sional space, DEA fits a piecewise linear envelope, or frontier, to the given data.
The envelope indicates a normative ideal, given the existing data. Points located
on the envelope are optimally efficient, while points below the envelope are inef-
ficient. DEA evaluates all points with respect to their deviation from the frontier.
The values of the points on the frontier equal to one and the values of other
points that operate beneath the frontier are between zero and one. DEA requires
identifying input and output variables. P. Demerjian et al. (2012) use seven input
variables: the cost of goods sold (COGS); selling, general, and administra-
tive expenses (XSGA); property, plant, and equipment (PPE); operating lease;
research, and development cost (R&D); goodwill; and other intangibles. The out-
put variable in P. Demerjian et al. (2012) is net sales.
Second, P. Demerjian et al. (2012) attempt to identify the manager-specific
characteristics of the total firm efficiency from DEA results because the total firm
efficiency can be attributed to both the manager-specific characteristics and firm-
specific characteristics. Thus, P. Demerjian et al. (2012) regress the total firm effi-
ciency on six firm-specific variables that could aid or hinder managers’ abilities.
These six variables include firm size, firm market share, cash available, firm age,
operational complexity, and foreign operations. This regression is run by indus-
try and with year-fixed effects to purge industry and year effects. P. Demerjian
et al. (2012) use the residuals from the regression as a proxy for managerial ability
because the residuals are attributed to managerial ability.
In addition, we adopt two alternative proxies for CEO ability used in the recent
literature in our robustness check. Our first alternative proxy for managerial abil-
ity (CEO_media_citation) is based on the number of press citations for a CEO
over a prior five-year period (Baik et al., 2009; Rajgopal et al., 2006).8 Our second
measure of CEO ability (CDF) follows Rajgopal et al. (2006) and uses the three-
year average of the CDF of ROA for each CEO-firm-year by industry.

3.3. Empirical Specification

To explore the relation between managerial ability and employee productivity, we
employ the following pooled cross-sectional models using ordinary least squares:

EMPLOYEEit = β0 + β1lag_MASCOREit-1 + β2SIZEit + β3MTBit

+ β4LEVit + β5ROAit + β6ASSETAGEit
+ β7LABORINTENit + β8ZSCOREit + β9EUit +
β10CAPINTENit + ΣINDUSTRYit + ΣYEARit + εit(5)

The dependent variable, EMPLOYEE, alternatively represents one of our three

employee measures: employee productivity (ln LP), employee efficiency (ln LE),
and employee cost (ln LC). The independent variable, MASCORE, represents the
managerial ability score developed from P. Demerjian et al. (2012). This measure
Managerial Ability and Employee Productivity 159

is calculated as the residual of P. Demerjian et al.’s (2012) step-two model. All

other control variables are defined in the Appendix. Petersen (2009) states that
the residuals of a given firm may be correlated across years (year/time effect), and
that the residuals of a given year may be correlated across different firms (firm
effect) in studies using panel data sets. To better control for the firm effects, we
calculate the robust standard errors with a two-way clustering by firm and year.
Also, to control for the year and industry effects on employee productivity, we
include the fixed effects for time and industry.
To test our hypotheses regarding employee productivity (ln LP) and employee
efficiency component (ln LE), we analyze the coefficient on MASCORE. To the
extent that more-able managers can manage their employees more efficiently, we
expect a positive and significant coefficient on MASCORE. To test our hypoth-
eses regarding the employee cost component (ln LC), we still analyze the coef-
ficient on MASCORE. To the extent that more-able managers can better control
their employee cost per employee, we expect a negative and significant coefficient
on MASCORE.
In addition to our variable of interest, we also include other variables that
are associated with managerial ability and employee variables established in the
prior literature. SIZE is calculated as the natural log of the total assets (AT)
and is used to control for the size of the firm on employee performance (Stuebs
& Sun, 2010). MTB is used to control a firm’s growth prospects. It is calcu-
lated as the ratio of the market value of the assets over the book value of the
assets. LEV is employed to proxy for the firm’s exposure to financial leverage
and is computed as the ratio of total long-term debt divided by AT. ROA is
used to measure a firm’s effectiveness in capturing business opportunities and
used as standard measure for firm performance. Additionally, we include the
age of long-term assets9 (ASSETAGE) in our analysis because the managers in
firms with older assets may be less responsive than those in firms with younger
assets (Cochran & Wood, 1984). We include labor intensity (LABORINTEN)
because employee productivity may vary in industries with different levels of
labor intensity. For example, employee productivity is typically lower in labor-
intensive industries, like restaurants, and is higher in high-tech industries. We
use ZSCORE to control for a firm’s financial health in their study because man-
agers behave differently when their firms are under financial stress (Huang &
Sun, 2015). We also include environmental uncertainty (EU) in our regression
analysis because we posit that managerial ability behaves differently in a more
volatile environment. Finally, we incorporate CAPINTEN to control for the
capital intensity of the firm.

4. RESULTS
4.1. Sample Selection and Descriptive Statistics
We begin our sample selection process by downloading the managerial abil-
ity scores from Dr. Sarah McVay’s website. This managerial ability sample has
194,186 firm-year observations from 1980 to 2013. Next, we use Compustat to
160 DIPANKAR GHOSH ET AL.

obtain financial statement data, which includes: AT; current assets (ACT); current
liabilities (LCT); long-term liabilities (DLTT); total net value of property, plant
and equipment (PPENT); total gross value of property, plant, and equipment
(PPEGT); outstanding common shares (CSHO); stock price at fiscal year-end
(PRCC_F); retained earnings (RE); sales (SALE); income before extraordinary
items (IB); NI; market value of equity (MKVLT); XLR; and EMP. We merge the
above two samples. Because few firms report employee cost (XLR), many obser-
vations are lost due to the lack of XLR.10 We then eliminate firms in the financial
(SIC 6000-6999) and utility industries (SIC 4000-4999) because those industries
are highly regulated. Our final sample with completed data consists of 8,568 firm-
year observations.
Table 1 reports the descriptive statistics for all the variables used in the pri-
mary analyses of our study. The mean (median) value of LP is 56.722 (39.769).
The mean (median) values of LE and LC are 1.386 (1.185) and 41.692 (33.253),
respectively. MASCORE has a mean value of 0.011. On average, our sample firms
have total assets (ASSETS) of $9,678 million and sales (SALE) of $8,660 million,
suggesting that our sample tends to be larger and more profitable firms. The
median values of ROA are 0.052, suggesting that a typical firm in our sample
earns a 5.2% profit on its assets. Overall, our sample does not exhibit high degree
of financial leverage (LEV), as our total long-term debt is only 17.6% of AT.
As expected, the majority of our sample has MTB greater than 1. In addition,
the average firm-year appears to be financially healthy as the mean of ZSCORE
is above 3.
Table 2 provides the correlation matrices for the main variables used in the
analyses. For each pair of variables, the Pearson and Spearman correlation coeffi-
cients are provided. Consistent with our hypotheses, both Pearson and Spearman
correlations indicate managerial ability (MASCORE) is positively correlated
with the employee productivity (ln LP) and employee efficiency (ln LE) compo-
nent at a significant level (p < 0.0001). Thus, the above findings offer descriptive
support for our hypotheses. In general, the employee variables (ln LP, ln LE, and
ln LC) are highly correlated with the control variables, including SIZE, MTB,
LEV, ROA, ASSETAGE, LABORINTEN, ZSCORE, EU, and CAPINTEN. For
example, EU is highly and significantly related to ln LP, ln LE, and ln LC, sug-
gesting EU has a negative impact on employee performance and cost.

4.2. H1 Testing
Tables 3 and 4 provide the results of our hypothesis test on the association between
managerial ability and employee productivity. These tests employ Equation (5)
and use employee productivity as the dependent variable. Table 3 serves as the
first test for our H1 by using the employee productivity measure from Stuebs and
Sun (2010) and the managerial ability proxy from P. Demerjian et al. (2012). Table
4 Panel A employs the alternative measure of employee productivity, while Panel
B employs the alternative measures of managerial ability.
Table 3 employs Equation (5) to evaluate the association between manage-
rial ability (lag_MASCORE) and employee productivity (ln LP). The t-statistics
 Table 1. Descriptive Statistics of Sample Firms.
N Mean Std. Dev. 10th %tile 25th %tile Median 75th %tile 90th %tile

LP 8,568 56.722 59.923 10.679 20.859 39.769 67.933 117.158

LE 8,568 1.386 0.948 0.856 1.057 1.185 1.413 1.915
LC 8,568 41.692 35.678 9.892 18.248 33.253 53.562 81.360
LnLP 8,568 3.605 0.988 2.368 3.038 3.683 4.219 4.764
LnLE 8,568 0.194 0.525 −0.156 0.055 0.170 0.346 0.650
LnLC 8,568 3.414 0.844 2.292 2.904 3.504 3.981 4.399
MASCORE 8,568 0.011 0.147 −0.126 −0.077 −0.024 0.050 0.208
CEO_media_citation 802 4.776 1.093 3.332 4.043 4.745 5.617 6.174
CDF 1,347 −2.009 34.609 −0.453 1.015 1.106 1.456 1.991
EMPPRO_ALT 1,280 −0.005 1.075 −1.000 −1.000 0.000 0.000 1.000
ASSETS 8,568 9,677.960 28,436.900 22.859 154.201 1,060.130 6,111.520 23,821.800
SALE 8,568 8,660.480 27,851.200 27.003 191.796 1,125.120 6,070.420 19,827.900
NI 8,568 584.289 2,157.620 −4.987 1.661 37.854 280.259 1,331.160
Managerial Ability and Employee Productivity

XLR 8,568 1,336.450 2910.350 6.859 45.617 291.869 1,333.260 3,578.410

EMP 8,568 32.602 62.881 0.320 1.958 9.703 38.283 84.500
SIZE 8,568 6.779 2.597 3.129 5.038 6.966 8.718 10.078
MTB 8,568 2.570 4.714 0.663 1.099 1.797 2.962 4.802
LEV 8,568 0.176 0.162 0.000 0.049 0.149 0.254 0.371
ROA 8,568 0.042 0.121 −0.038 0.019 0.052 0.087 0.127
ASSETAGE 8,568 0.538 0.160 0.340 0.437 0.539 0.641 0.737
LABORINTEN 8,568 0.290 0.356 0.091 0.165 0.255 0.353 0.487
ZSCORE 8,568 3.485 5.438 1.070 1.890 2.898 4.430 6.578
EU 8,568 0.737 0.578 0.214 0.354 0.588 0.935 1.430
CAPINTEN 8,568 0.071 0.063 0.016 0.031 0.056 0.093 0.140

Note: Please see the Appendix for variable definitions.

161
 Table 2. Correlations Among Selected Variables (1980–2013).
LnLP LnLE LnLC MAS- CEO_ CDF EMPPRO_ SIZE MTB LEV ASSE- LABOR- ZSCORE EU CAP- LP LE LC
CORE media_ ALT TAGE INTEN INTEN
citation

LnLP 0.448 0.872 0.241 0.310 –0.186 0.192 0.486 0.229 –0.066 –0.234 –0.160 0.053 –0.187 –0.190 1.000 0.448 0.872
LnLE 0.521 0.078 0.273 0.035 –0.029 0.174 0.330 0.339 –0.136 0.122 –0.514 0.415 0.002 0.128 0.448 1.000 0.078
LnLC 0.841 –0.0011 0.158 0.347 –0.179 0.171 0.405 0.136 –0.022 –0.335 0.038 –0.124 –0.207 –0.279 0.872 0.078 1.000
MASCORE 0.285 0.229 0.194 0.151 –0.004 0.211 0.169 0.160 –0.205 –0.155 –0.273 0.231 –0.016 0.031 0.241 0.273 0.158
CEO_media_ 0.251 –0.002 0.278 0.191 –0.088 –0.199 0.643 0.160 0.099 –0.173 –0.117 –0.204 –0.172 –0.214 0.310 0.0035 0.347
citation
CDF –0.014 –0.010 –0.010 0.003 0.015 –0.190 –0.083 –0.017 –0.015 0.111 0.007 0.009 0.000 0.159 –0.186 –0.029 –0.179
CSR_Employee 0.235 0.083 0.209 0.217 –0.220 –0.080 0.170 0.132 –0.097 –0.086 –0.119 0.064 –0.012 –0.076 0.192 0.174 0.171
SIZE 0.458 0.274 0.361 0.296 0.643 0.019 0.217 0.157 0.222 –0.010 –0.303 –0.141 –0.219 0.067 0.486 0.330 0.405
MTB 0.069 0.061 0.043 0.067 0.086 0.018 0.054 0.020 –0.063 0.016 –0.034 0.393 –0.013 0.065 0.229 0.339 0.136
LEV –0.087 –0.121 –0.028 –0.169 0.087 0.032 –0.107 0.110 –0.012 0.194 –0.047 –0.474 –0.041 0.087 –0.066 –0.136 –0.022
ASSETAGE –0.171 0.141 –0.289 –0.100 –0.152 0.027 –0.088 0.056 0.006 0.166 –0.073 0.053 0.268 0.327 –0.234 0.122 –0.335
LABORINTEN –0.005 –0.161 0.089 –0.158 –0.024 0.008 –0.043 –0.169 0.000 –0.036 –0.071 –0.081 –0.052 –0.136 –0.160 –0.514 0.038
ZSCORE 0.071 0.233 –0.056 0.058 –0.185 0.003 0.040 –0.033 0.153 –0.271 0.132 –0.044 0.050 0.171 0.052 0.415 –0.124
EU –0.176 –0.034 –0.182 –0.032 –0.117 0.035 –0.046 –0.242 0.043 0.006 0.253 0.009 0.019 0.050 –0.187 0.002 –0.207
CAPINTEN –0.147 0.087 –0.222 0.029 –0.244 0.018 –0.078 –0.046 0.044 0.046 0.307 –0.076 0.106 0.134 –0.190 0.128 –0.279
LP 0.798 0.476 0.648 0.309 0.227 0.017 0.259 0.345 0.075 –0.086 –0.110 0.002 0.057 –0.097 –0.092 0.448 0.872
LE 0.389 0.788 –0.037 0.226 –0.027 0.001 0.087 0.185 0.052 –0.111 0.124 –0.131 0.154 0.027 0.066 0.535 0.078
LC 0.697 0.006 0.836 0.188 0.284 0.018 0.192 0.253 0.038 –0.050 –0.268 0.126 –0.028 –0.112 –0.186 0.743 0.045

Notes: Pearson correlation is below, and the Spearman correlation is above the diagonal. Please see the Appendix for variable definitions. Total observations = 8,568.
Managerial Ability and Employee Productivity 163

Table 3. Regression Analysis.

Testing H1: Managerial Ability and Employee Productivity.
Managerial ability is positively related to employee productivity (H1).
(1) LnLP

lag_MASCORE 0.159***
(5.460)
SIZE 0.127***
(6.520)
MTB 0.002
(1.190)
LEV 0.029
(0.420)
ROA 2.109***
(9.040)
ASSETAGE −0.432***
(−5.450)
LABORINTEN 0.228***
(7.970)
ZSCORE −0.003
(−0.930)
EU −0.046**
(−2.300)
CAPINTEN −0.107
(−0.660)
INDUSTRY F.E.’s Yes
YEAR F.E.’s Yes
INTERCEPT Yes
N 8,568
R2 0.531

Notes: Please see the Appendix for variable definitions. Our VIF analysis suggests that multicollinearity
is not an issue (untabulated).
***p < 0.01, **p < 0.05, *p < 0.1.

are calculated using firm-level cluster robust standard errors. Overall, the results
show a positive association between lag_MASCORE and ln LP, which suggests
that a higher ability manager is associated with better employee performance.
This finding supports our H1. In Table 3, we find that the coefficient on lag_
MASCORE is positive (0.159) and highly significant (t-statistics = 5.460). This
result seems to be practically significant, as well. A one-standard deviation change
in lag_MASCORE (0.147) results in a 1.024 (0.147 × 0.159 exponential) change
in employee productivity. In addition, many of the control variables exhibit a
significant relation with ln LP as expected. SIZE and ROA are positively related
to ln LP, which is consistent with the intuition that larger firms and firms operate
effectively have better employee productivity. Firms with higher labor intensity
show better employee productivity, while firms with older assets are associated
with lower employee productivity. Also, firms operate under high EU appear to
have low employee productivity.
164 DIPANKAR GHOSH ET AL.

Table 4. Regression Analysis.

Panel A: Testing H1 Using Alternative Employee Productivity Proxy – EMPPRO_ALT
(1) EMPPRO_ALT
lag_MASCORE 0.180*
(1.680)
SIZE 0.091**
(2.060)
MTB 0.007*
(1.770)
LEV −0.675***
(−3.940)
ROA 0.325
(0.890)
ASSET AGE 0.028
−0.12
LABORINTEN −0.022
(−0.730)
ZSCORE 0.004
(0.280)
EU 0.002
(0.030)
CAPINTEN 0.031
(0.050)
INDUSTRY F.E.’s Yes
YEAR F.E.’s Yes
INTERCEPT Yes
N 1,335
R2 0.28

Panel B: Testing H1 Using Alternative Managerial Ability Proxies – CEO_media_mention and CDF

(1) LnLP (2) LnLP

CEO_media_citation 0.035*
(1.710)
CDF 0.014*
(1.850)
SIZE 0.092 0.149**
(1.190) (2.220)
MTB 0.010** 0.000
(2.330) (0.570)
LEV −0.037 0.034
(−0.280) (0.310)
ROA 4.018*** 3.708***
(6.990) (7.360)
ASSETAGE −0.670*** −0.011
(−2.800) (−0.060)
LABORINTEN 0.042 0.087
(0.640) (1.590)
ZSCORE −0.015** −0.019**
(−2.220) (−2.390)
Managerial Ability and Employee Productivity 165

(1) LnLP (2) LnLP

EU −0.038 0.023
(−0.9200) (0.600)
CAPINTEN −0.005 −0.024
(−0.010) (−0.050)
INDUSTRY F.E.’s Yes Yes
YEAR F.E.’s Yes Yes
INTERCEPT Yes Yes
N 803 1,231
R2 0.776 0.733

Note: Please see Appendix for variable definitions

***p < 0.01, **p < 0.05, *p < 0.1.

Besides, we conduct two of robustness tests. Our first robustness test involves
the dependent variable in our main analysis in Equation (5). More specifically,
we use EMPPRO_ALT instead of ln LP as our proxy for employee productiv-
ity. EMPPRO_ALT is the employee performance ratings provided by the MSCI’s
ESG database. Table 4 Panel A reports the results testing H1 using this alternative
measure of employee productivity. We find similar results as the main analysis:
more-able managers are associated with higher employee productivity. The coeffi-
cient on lag_MASCORE is 0.180 and statistically significant (t-statistics = 1.680).
Our second set of robustness tests involves the primary independent variable
in our analysis. We use two alternative measures, namely CEO_media_citation
and CDF, to proxy for managerial ability. Table 4 Panel B shows the results test-
ing our H1 using these two alternative proxies of managerial ability. We find that
both measures of managerial ability appear to be positively related to employee
productivity at a significant level. The results further confirm our main findings
in Table 3.11 Taken together, results of Table 4 indicate that our primary findings
are robust to alternative employee productivity and managerial ability measures,
strengthening our main hypothesis (H1).

4.3. H2a and H2b Testing

The previous subsection presents the baseline findings that firms with more-able
managers are more likely to demonstrate higher employee productivity. In this
section, we further decompose the employee productivity measure into two com-
ponents (the employee efficiency component and the employee cost component)
and test which component drives the employee productivity result. We employ
Equation (5) while using the employee efficiency or the employee cost measure as
the dependent variable. Table 5 reports the results based on Equation (5).
Column 1 uses the employee efficiency measure (ln LE) as the dependent vari-
able, while Column 2 uses the employee cost measure (ln LC). The sample in
Columns 1 and 2 is the overall sample of 8,568 firm-years. In Column 1, the coef-
ficient on lag_MASCORE is positive (0.130) and highly significant (t-statistics =
6.340). This result suggests that more-able managers are more likely to have
166 DIPANKAR GHOSH ET AL.

Table 5. Regression Analysis.

Testing H2a and H2b: Managerial Ability and the
Employee Efficiency/Cost Component.
(1) LnLE (2) LnLC

lag_MASCORE 0.130*** −0.017

(6.340) (−0.660)
SIZE 0.194*** −0.153***
(19.800) (−10.110)
MTB 0.000 0.001
(0.280) (0.660)
LEV −0.122** 0.158***
(−2.420) (3.340)
ROA 2.465*** −0.148***
(10.420) (−3.890)
ASSETAGE 0.093** −0.451***
(2.080) (−6.630)
LABORINTEN 0.049*** 0.112***
(2.640) (7.010)
ZSCORE −0.001*** 0.003**
(−2.800) (2.190)
EU −0.005 −0.009
(−0.360) (−0.570)
CAPINTEN 0.278*** −0.356**
(2.630) (−2.370)
INDUSTRY F.E.’s Yes Yes
FIRM FIXED EFFECTS Yes Yes
INTERCEPT Yes Yes
N 8,568 8,568
R2 0.538 0.481

Note: Please see the Appendix for variable definitions.

***p < 0.01, **p < 0.05, *p < 0.1.

higher employee efficiency. This result confirms H2a. In Column 2, the coefficient
on lag_MASCORE is negative (−0.017) but insignificant (t-statistics = −0.660)
from a statistical standpoint, which does not support our H2b. Taken together,
we find results in Table 5 suggest that the positive relation between managerial
ability and employee productivity is possibly driven by better employee efficiency,
but not necessarily by the lower employee cost.

5. ADDITIONAL TESTS
5.1. Financial distress
Financial distress indicates a condition when payments are not made or made
with difficulty; that is, it is reasonably unlikely that the company will be able to
pay all of its debts as they fall due. When in financial distress, companies are
Managerial Ability and Employee Productivity 167

often confronted with the dilemma of raising capital to fund their restructuring,
especially if a financial boost is not a guarantee to provide a lasting solution to
the problems at hand. Thus, it is agreed that managers must have specific skills
and knowledge and that they must be adept at handling the unique challenges
that exist in financially distressed firms. Although the skills used in managing in a
distressed situation – leadership, decisiveness, insightfulness, and motivation – are
not different from those used in managing a firm that is not in financial distress,
their use and application are much different than with managing firms that are
not in financial distress. Managers are maneuvering closer to the ground, so there
is a lesser margin for error, and it is merely a lot harder to lead well when cash
flow is tight (Kim, 2012).
We use ZSCORE to measure the financial health of a firm. We test the
interaction term lag_MASCORE × ZSCORE to evaluate whether the relation
between financial distress and employee productivity varies across managerial
ability levels. Table 6 provides the results for this test. As predicted, we find a
significant positive relation (coefficient = 0.124 and t-statistics = 1.940) between

Table 6. Regression Analysis.

Testing H1 across Financial Distress Levels.
(1) LnLP

lag_MASCORE 0.124*
(1.940)
Zscore −0.009
(−1.070)
lag_MASCORE x Zscore 0.011
(0.660)
SIZE 0.081***
(17.370)
MTB 0.002
(1.070)
LEV 0.030
(0.430)
ROA 2.127***
(9.110)
ASSETAGE −0.383***
(−4.860)
LABQRINTEN 0.197***
(8.530)
EU −0.037*
(−1.900)
CAPINTEN −0.149
(−0.920)
INDUSTRY F.E.s Yes
YEAR F.E.’s Yes
INTERCEPT Yes
N 8,568
R2 0.531

Note: Please see the Appendix for variable definitions.

***p < 0.01, **p < 0.05, *p < 0.1.
168 DIPANKAR GHOSH ET AL.

lag_MASCORE and ln LP, which suggests that for an average financial distress
firm, the higher managerial ability is associated with higher employee produc-
tivity. However, we fail to find any relation between ZSCORE and ln LP as the
coefficient is negative but not statistically significant. Similarly, we do not find
significant relation between the interaction term lag_MASCORE × ZSCORE
and ln LP, which suggests that managerial ability does not appear to have dif-
ferent impacts on employee productivity upon different levels of firm financial
distress.

5.2. Environmental Uncertainty

EU refers to the degree, or variability, of change that characterizes environ-
mental activities relevant to an organization’s operations (Govindarajan, 1988).
Managers have difficulty in predicting or attaching probabilities to a firm’s EU
factors, for they are constantly changing (e.g., actions of the customers, suppli-
ers, competitors, and regulatory groups) (Dess & Beard, 1967; Drago, 1998).
However, managers can make strategic choices (Child, 1972); that is, although
firms are constrained by their environments, managers have some latitude to act,
which translates to fundamentally different ways of managing and controlling
recourses via the design of management control system (MCS) (Bowyer, 1970;
Govindarajan, 1988; Simons, 1995). Since resources are affected by the environ-
mental context in which the firm operates (Lichtenstein & Brush, 2001), to fully
understand the linkage between the management of resources and the creation of
value, the effects of a firm’s external environment on managing resources need to
be examined (Bettis & Hitt, 1995).
Early management and accounting literature (Child, 1972; Milliken, 1987;
Kren, 1992) consistently suggest that market characteristics are the most appro-
priate measures for a firm’s environment. Following prior studies (e.g., Dess &
Beard, 1984; Ghosh & Olsen, 2009; Snyder & Glueck, 1982; Tosi, Aldag, &
Storey, 1973), we use the coefficient of variation (CV) of sales to capture EU. The
formula is expressed below:
2
(Si − Smean )
∑
5
i =1 5
CV (Si ) =
Smean

where Si is a firm’s sales in the year i and Smean is the mean of sales over five years.
For example, if we calculate a CV of sales for firm X in 2015, Smean is the mean of
sales for the period of 2010–2014 for firm X. We calculate CV of sales by year and
industry (the first two-digit of SIC codes). Following Ghosh and Olsen (2009), we
normalize the raw firm-specific EU by dividing it by the average EU for that firm’s
industry for the same year to mitigate industry effects. A higher value of CV of
sales indicates a higher level of EU.
To test the moderating effect of managerial ability on the relation between EU
and employee productivity, we test the interaction term of lag_MASCORE ×
EU. Table 7 reports the results for this test. We find that EU is negatively and
Managerial Ability and Employee Productivity 169

Table 7. Regression Analysis.

Testing H1 across Environmental Uncertainty Levels.
(1) LnLP

lag_MASCORE 0.006
(0.120)
EU −0.156***
(−3.700)
lag_MASCORE x EU 0.207***
(3.260)
SIZE 0.083***
(17.480)
MTB 0.002
(1.030)
LEV 0.030
(0.450)
ROA 2.145***
(9.070)
ASSETAGE −0.383***
(−4.940)
LABORINTEN 0.197***
(8.630)
ZSCORE −0.003
(−0.920)
CAPINTEN −0.175
(−1.090)
INDUSTRY F.E.’s Yes
YEAR F.E.’s Yes
INTERCEPT Yes
N 8,568
R2 0.532

Note: Please see the Appendix for variable definitions.

***p < 0.01, **p < 0.05, *p < 0.1.

significantly (coefficient = −0.156 and t-statistics = −3.700) related to ln LP. This

finding suggests that for a firm with an average ability manager, the higher the
EU, the firm has lower employee productivity. In addition, we find a positive and
statistically significant relation between the interaction term lag_MASCORE ×
EU and ln LP (coefficient = 0.207 and t-statistics = 3.260). This finding suggests
that managerial ability moderates the negative effect of uncertain environments
on employee productivity.

5.3. Technology Firms

Because not all firms generate income from employees in a similar manner, we perceive
that the influence of the industry membership on employee productivity would be
moderated by managerial ability. Following Kile and Phillips (2009), we use Industry
Classification Codes (SIC) to classify our sample into high-technology firms and low-
technology firms. The firms with SIC starts with 283 (Drugs), 357 (Computer and
170 DIPANKAR GHOSH ET AL.

Office Equipment), 366 (Communication Equipment), 367 (Electronic Components

and Accessories), 382 (Laboratory, Optic, Measure, Control Instruments), 384
(Surgical, Medical, Dental Instruments), 481 (Telephone Communications), 482
(Miscellaneous Communication Services), 489 (Communication Services, NEC), 737
(Computer Programming, Data Processing, etc.), and 873 (Research, Development,
Testing Services) are classified as high-technology firms, while the rest of the firms in
our sample are classified as low-technology firms.
We report our results in Table 8. We code the variable HItech equals to 1 if
the firm falls within high-technology industries (based on the SIC code) as dis-
cussed above, and equals to 0 otherwise. Our high-technology firm sample has
1,321 firm-years, while our low-technology sample has 7,247 firm-years. On
the one hand, Table 8 shows a strong positive relation (coefficient = 0.114 and
t-statistics = 3.620) between lag_MASCORE and ln LP. This finding suggests

Table 8. Regression Analysis.

Testing H1 in High-technology Firms and Low-technology Firms.
(1) LnLP

lag_MASCORE 0.114***
(3.620)
HItech −0.113*
(−1.720)
lag_MASCORE x HItech 0.283***
(3.870)
SIZE 0.081***
(16.680)
MTB 0.002
(0.940)
LEV 3.030
(0.440)
ROA 2.148***
(9.270)
ASSETAGE −0.379***
(−4.880)
LABORINTEN 0.198***
(8.660)
ZSCORE −0.003
(−0.980)
EU −0.035*
(−1.810)
CAPINTEN −0.147
(−0.910)
INDUSTRY F.E.’s Yes
YEAR F.E.’s Yes
INTERCEPT Yes
N 8,568
R2 0.531

Note: Please see the Appendix for variable definitions.

***p < 0.01, **p < 0.05, *p < 0.1.
Managerial Ability and Employee Productivity 171

that for low-technology firms, high managerial ability has a positive effect on
employee productivity. On the other hand, we find a negative and marginally
statistically significant relation between HItech and ln LP (coefficient = −0.113
and t-statistics = −1.720). This finding suggests that on average high-technology
firms have marginally lower employee productivity than low-technology firms.
This finding makes sense when we consider the higher pay and longer return
cycle in the high-technology firms. Also, Table 8 yields interesting results regard-
ing the interaction term lag_MASCORE × HItech. We find a significantly posi-
tive coefficient on lag_MASCORE × HItech (coefficient = 0.283 and t-statistics
= 3.870), suggesting that the negative association between high-technology firms
and employee productivity is moderated by higher ability managers.

5.4. Capital Intensity

Prior research suggests that capital intensity is one of the factors causing different
levels of employee productivity. In this subsection, we test whether managerial
ability moderates the relation between capital intensity and employee produc-
tivity. To test this question, we examine the coefficient on the interaction term
Lag_MASCORE × CAPINTEN. The results fail to find the moderating effect
of managerial ability as the coefficient on the interaction term is positive but
not statistically significant. However, the significant positive coefficient on Lag_
MASCORE (coefficient = 0.133 and t-statistics = 2.750) confirms the positive
relation between managerial ability and employee productivity (Table 9).

5.5. Financial Crisis

In this test, we examine whether macroeconomic factors such as the financial
crises can influence our primary findings. We divide our sample period into three
periods: before financial crisis (2004–2007), during financial crisis (2008–2009),
and after financial crisis (2010–2013). We identify 1,105 firm-years in our “before
crisis” subsample, 531 firm-years in our “during crisis” subsample, and 1,151
firm-years in our “after crisis” subsample. As shown in Table 10, coefficients on
lag_MASCORE are significantly positive in these three subsamples. These find-
ings suggest that more-able managers lead to higher employee productivity before
the financial crisis, during the financial crisis, and after the financial crisis. In
other words, our initial results are persistent over time.

6. CONCLUSION
This study demonstrates that managers are significant sources of value crea-
tion. Without considering managerial ability in employee performance research,
one assumes homogeneity across firms, which we demonstrate is not a realistic
assumption. Furthermore, we suggest that future research on employee perfor-
mance should pay attention to the managerial ability of the organizations, rather
than focusing solely on organization-level characteristics. As our results indicate,
managers differ in their abilities to synchronize management processes, human
172 DIPANKAR GHOSH ET AL.

Table 9. Regression Analysis.

Testing H1 across Capital Intensity Levels.
(1) LnLP

lag_MASCORE 0.133***
(2.750)
CAPINTEN −0.389
(−1.050)
lag_MASCORE x CAPINTEN 0.387
(0.690)
SIZE 0.082***
(17.190)
MTB 0.002
(1.070)
LEV 0.033
(0.490)
ROA 2.143***
(9.170)
ASSETAGE −0.381***
(−4.890)
LABORINTEN 0.195***
(8.640)
ZSCORE −0.003
(−0.900)
EU −0.036*
(−1.860)
INDUSTRY F.E.’s Yes
YEAR F.E.’s Yes
INTERCEPT Yes
N 8,568
R2 0.530

Note: Please see the Appendix for variable definitions.

***p < 0.01, **p < 0.05, *p < 0.1.

capital, and other resources in ways that enhance employee productivity. We also
find a substantial connection between managerial ability and firms’ operating
circumstances, as managerial ability becomes increasingly important in firms that
are in high-technology industry or that are faced with high EU, which suggests
that the influence of able managers on employee productivity depends on the
firm’s circumstances.
Our study has practical implications. Our results suggest that managers do
matter in employee performance. Hence, investors or shareholders should con-
sider the managerial ability before they make investment decisions because firms
with more-able managers are more productive than firms with less-able managers.
Additionally, board of directors should hire and retain more-capable managers
because these managers can better manage the most valuable asset – employees.
Our results may also encourage managers to improve their abilities, which can not
only secure their positions but also improve their firms’ operating performance. To
employees, our findings may encourage them to work for more-capable managers.
Managerial Ability and Employee Productivity 173

Table 10. Regression Analysis.

Testing H1 around the Financial Crisis Period.
(1) Before Crisis (2004- (2) During Crisis (2008- (3) After Crisis (2010-
2007) DV = LnLP 2009) LnLP 2013) LnLP

lag_MASCORE 0.268*** 0.337*** 0.489***

(3.460) (3.020) (6.010)
SIZE 0.084* 0.055 0.092**
(1.870) (0.750) (2.110)
MTB 0.004 0.003 0.006
(0.750) (0.440) (1.340)
LEV 0.246 0.011 0.337**
(1.600) (0.060) (2.290)
ROA 3.283*** 1.963*** 1.997***
(7.830) (3.340) (3.330)
ASSETAGE −0.368* −0.604* −0.170
(−1.680) (−1.890) (−0.740)
LABORINTEN 0.143* 0.151 0.167***
(1.860) (1.540) (6.680)
ZSCORE −0.012 −0.017 −0.002
(−1.570) (−1.390) (−0.320)
EU 0.038 −0.003 −0.125**
(0.650) (−0.040) (−2.380)
CAPINTEN −0.828 −2.284*** −0.463
(−1.570) (−3.000) (−0.650)
INDUSTRY F.E.’s Yes Yes Yes
YEAR F.E.’s Yes Yes Yes
INTERCEPT Yes Yes Yes
N 1,105 531 1,151
R2 0.600 0.527 0.473

Note: Please see the Appendix for variable definitions.

***p < 0.01, **p < 0.05, *p < 0.1.

This study has its limitations. First, it is hard to measure managerial ability
because it is multidimensional. The managerial ability scores by P. Demerjian
et al. (2012) are an approximate measure of management performance. More pre-
cise measures of management performance may yield stronger results. Second, our
sample size is relatively small due to the limited data available on employee cost
(XLR) in Compustat. Besides, due to these data availability issue, our sample tends
to capture larger firms in general. Lastly, our study may raise concerns about omit-
ted variables in our analyses. For example, variables such as employee ability and
motivation are not included. Readers need to exercise caution when generalizing
the conclusions. The above issues can be investigated in future studies.

NOTES
1. For example, Holcomb et al. (2009) show that managerial ability affects resource
productivity, although this effect diminishes as the quality of the resources increase. The
authors define three types of managerial ability: general ability, firm-specific ability, and
174 DIPANKAR GHOSH ET AL.

industry-specific ability. Using data from a single industry (i.e., National Football League),
the authors show that firm- and industry-specific sources of managerial ability are linked
to better organizational performance. However, because they emphasize one industry,
their paper only focuses on firm- and industry-specific abilities. As we discuss later, using
P. Demerjian et al.’s (2012) managerial ability measure that is a performance-based meas-
ure of managers’ efficiency in transforming corporate resources to revenues and our cross-
industry sample, our study complements their study by showing that managers’ general
ability is also linked to organizational performance (i.e., employee productivity). Our
measures of managerial ability as well as employee productivity are more direct (hence,
less likely to have measurement errors). In addition, since we examine firm-level impact of
managerial ability encompassing a broader cross-section of industries, our results are more
generalizable.
2. We appreciate Professor Demerjian for his generosity to share this DEA measure of
managerial ability on his website.
3. We appreciate Professor Baik for his generosity to share these media citation data
with us.
4. This view of managerial ability as consisting of knowledge, skills, and experience
embodied within an individual is largely consistent with prior descriptions of human capi-
tal (Becker, 1964)
5. Compustat Item #18.
6. Compustat Item #42.
7. Compustat Item #29.
8. We appreciate Professor Baik for his generosity to share these media citation data
with us.
9. The ratio of net long-term assets to gross long-term assets is used to measure asset
age. The newer a firm’s long-term assets, the closer this ratio will be to unity. As a firm’s
long-term assets age, this ratio will approach zero.
10. Due to these strict sample selection criteria, our sample size reduced from 15,033
firm-year to 8,568 firm-year. We further compare the difference in these two samples to
reveal the implication of this sample selection criteria and reveal the limitation of our find-
ings. Based on our comparison, we find that our sample firm-year on average is larger (AT),
has higher sales (SALE) and net income (NI), shows better ROA, and is less likely to fall
into financial distress (ZSCORE) than the general Compustat population.
11. Note that due to the data limitations, our sample size drops from 8,568 firm-year to
803 firm-year and 1,231 firm-year using CEO_media_citation and CDF, respectively. Simi-
larly, our sample size drops to 1,335 firm-year if we use EMPPRO_ALT as the proxy for
employee productivity. To avoid potential bias due to this much smaller sample size, in the
following sections, we only report regression results using our main proxy for managerial
ability and employee productivity.

REFERENCES
Abernethy, M., & Chua, W. (1996). A field-study of control system “redesign”: The impact of institu-
tional processes on strategic choice. Contemporary Accounting Research, 13(2), 569–696.
Adner, R., & Helfat, C. (2003). Corporate effects and dynamic managerial capabilities. Strategic
Management Journal, 24(10), 1011–1025.
Aier, J., J. Comprix, Gunlock, M., & Lee, D. (2005). The financial expertise of CFOs and accounting
restatements. Accounting Horizon, 19(3), 123–135.
Alchian, A. A., & Demsetz, H. (1972). Production, information costs, and economic organization.
American Economic Review, 62(5), 777–795.
Anthony, R. (1965). Planning and control systems: A framework for analysis. Boston, MA: Harvard
Business, Division of Research.
Baik, B., David, B. F., & Lee, S. (2011). CEO ability and management earnings forecasts. Contemporary
Accounting Research, 28(5), 1645–1668.
Managerial Ability and Employee Productivity 175

Bailey, E., & Helfat, C. (2003). External management succession, human capital, and firm perfor-
mance: an integrative analysis. Managerial and Decision Economics, 24(4), 347–369.
Bamber, L. J., & Wang, J. I. (2010). What’s my style? The influence of top managers on voluntary
­corporate financial disclosure. The Accounting Review, 85(4), 1131–1162.
Barker, V. L., & Duhaime, I. M. (1997). Strategic change in the turnaround process: Theory and empiri-
cal evidence. Strategic Management Journal, 18(1), 13–38.
Barney, J. B. (1991). Firm resources and sustained competitive advantage. Journal of Management,
17(1), 99–120.
Baum, R., Locke, E., & Smith, K. (2001). A multi-dimensional model of venture growth. Academy of
Management Journal, 44, 292–303.
Becker, G. S. (1964). Human capital. New York, NY: Columbia University Press.
Berk, J., Stanton, R., & Zechner, J. (2010). Human capital, bankruptcy, and capital structure. Journal
of Finance, 65, 891–926.
Bertrand, M., & Schoar, A. (2003). Managing with style: The effect of managers on firm policies.
Quarterly Journal of Economics, 118, 1169–1208.
Bettis, R. A., & Hitt, M. A. (1995). The new competitive landscape. Strategic Management Journal,
16, 7–19.
Bowyer, J. (1970). Managing the resource allocation process. Cambridge, MA: Harvard University Press.
Butler, S. A., & Ghosh, D. (2015). Individual differences in managerial accounting judgments and decision
making. The British Accounting Review, 47(1), 33–45.
Campbell, D. (2012). Employee selection as a control system. Journal of Accounting Research, 50(4),
931–966.
Carlin, B., & Gervais, S. (2009). Work ethics, employment contracts, and firm value. Journal of Finance,
64, 785–821.
Carpenter, M. A., Sanders, W. G., & Gregersen, H. B. (2001). Bundling human capital with organiza-
tional context: The impact of international assignment experience on multinational firm perfor-
mance and CEO pay. Academy of Management Journal, 44(3), 493–511.
Castanias, R. P., & Helfat, C. E. (1991). Managerial resources and rents. Journal of Management, 17(1),
155–171.
Castanias, R. P., & Helfat, C. E. (2001). The managerial rents model: Theory and empirical analysis.
Journal of Management, 27(6), 661–678.
Chenhall, R. H. (2003). Management control systems design within its organizational context: find-
ing from contingency studies and directions for future research. Accounting, Organizations &
Society, 28(2/3), 127–168.
Child, J. (1972). Organizational structure, environment and performance: The role of strategic choice.
Sociology, 6, 2–22.
Cochran, R., & Wood, R. (1984). Corporate social responsibility and financial performance. Academy
of Management Journal, 27(1), 42–56.
Coff, R. W. (1997). Human assets and management dilemmas: Coping with hazards on the road to
resource-based theory. Academy of Management Review, 22(2), 374–402.
Collins, J., Holcomb, T. R., Certo, S. T., Hitt, M. A., & Lester, R. H. (2009). Learning by doing: Cross-
border mergers and acquisitions. Journal of Business Research, 62(12), 1329–1334.
Davis, S., & Albright, T. (2004). An investigation of the effect of balanced scorecard implementation on
financial performance. Management Accounting Research, 15, 135–153.
Dearman, D. T., & Shields, M. D. (2005). Reducing accounting fixation: Determinants of cogni-
tive adaptation to variation in accounting method. Contemporary Accounting Research, 22,
351–384.
Demerjian, P., Lev, B., Lewis, M., & McVay, S. (2013). Managerial ability and earnings quality. The
Accounting Review, 88(2), 463–498.
Demerjian, P., Lev, B., & McVay, S. (2012). Quantifying managerial ability: A new measure and validity
tests. Management Science, 58(7), 1229–1248.
Demerjian, P. R., Lewis-Western, M. F., & McVay, S. (2015). Earnings smoothing: for good or evil?
Retrieved from SSRN: http://ssrn.com/abstract=2426313
Dess, G., & Beard, D. (1984). Dimensions of organizational environments. Administrative Science
Quarterly, 29, 313–327.
176 DIPANKAR GHOSH ET AL.

Dierynck, B., Landsman, W. R., & Renders, A. (2012). Do managerial incentives drive cost behavior?
Evidence about the role of the zero earnings benchmark for labor cost behavior in private Belgian
firms. The Accounting Review, 87(4), 1219–1246.
Drago, W. A. (1998). Predicting organizational objectives: Role of stakeholder influence and uncertainty
of environmental sectors. Management Research News, 21(9), 16–28.
Dyreng, S., Hanlon, M., & Maydew, E. (2010). The effects of executives on corporate tax avoidance.
The Accounting Review, 85(4), 1163–1189.
Edmans, A. (2011). Does the stock market fully value intangibles? Employee satisfaction and equity
prices. Journal of Financial Economics, 101(3), 621–640.
Ezzamel, M. (1990). The impact of environmental uncertainty, managerial autonomy and size on
budget characteristics. Management Accounting Research, 1, 181–197.
Francis, J., Nanda, D., & Olsson, P. (2008). Voluntary disclosure, earnings quality, and cost of capital.
Journal of Accounting Research, 46(1), 53–99.
Ge, W., Matsumoto, D., & Zhang, J. (2011). Do CFOs have style? An empirical investigation of the
effect of individual CFOs on accounting practices. Contemporary Accounting Research, 28(4),
1141–1179.
Ghosh, D., & Olsen, L. (2009). Environmental uncertainty and managers’ use of discretionary accruals.
Accounting, Organizations & Society, 34(2), 188–205.
Ghosh, D., & Willinger, G. L. (2012). Management control systems, environmental uncertainty, and
organizational slack: Empirical Evidence. Advances in Management Accounting, 21, 87–118.
Govindarajan, V. (1988). A contingency approach to strategy implementation at the business-unit level:
Integrating administrative mechanisms with strategy. Academy of Management Journal, 31(4),
828–853.
Hambrick, D. (2007). Upper echelons theory: An update. Academy of Management Review, 32(2), 334–343.
Hambrick, D. C., & Mason, P. (1984). Upper Echelons: The organization as a reflection of its top
managers. Academy of Management Review, 9, 193–206.
Hatch, N. W., & Dyer, J. H. (2004). Human capital and learning as a source of sustainable competitive
advantage. Strategic Management Journal, 25, 1155–1178.
Helfat, C., & Peteraf, M. (2015). Managerial cognitive capabilities and the microfoundations of
dynamic capabilities. Strategic Management Journal, 36(6), 831–850.
Henri, J. F. (2006). Management control systems and strategy: A resource-based perspective. Accounting,
Organizations & Society, 31(6), 529–558.
Hitt, M. A., Bierman, L., Shimizu, K., & Kochhar, R. (2001). Direct and moderating effects of human
capital on strategy and performance in professional service firms: A resource-based perspective.
Academy of Management Journal, 44(1), 13–28.
Holcomb, T. R., Holmes, R. M., & Connelly, B. L. (2009). Making the most of what you have: manage-
rial ability as a source of resource value creation. Strategic Management Journal, 30, 457–485.
Holcomb, T. R., Holmes, R. M., & Hitt, M. A. (2006). Diversification to achieve scale and scope: The
strategic implications of resource management for value creation. In J. A. C. Baum, S. Dobrev, &
A. van Witteloostuijin (Eds.), Advances in strategic management: ecology and strategy (Vol. 23,
pp. 549–587). Oxford: JAI-Elsevier Science.
Hopper, T., & Armstrong, P. (1991). Cost accounting, controlling labor and the rise of conglomerates.
Accounting, Organizations and Society, 16(5–6), 405–438.
Huang, S. X., & Sun, L. (2015). The role of managerial ability on the relation between real earnings man-
agement and future firm performance. Working Paper. Ball State University.
Ittner, C. D., & Larcker, D. F. (2001). Assessing empirical research in managerial accounting: A value-
based management perspective. Journal of Accounting and Economics, 32, 349–410.
Ittner, C., Larcker, D., & Randall, T. (2003). Performance implications of strategic performance meas-
urement in financial services firms. Accounting, Organizations & Society, 28(7/8), 715–741.
Jensen, M. C. (1989). The eclipse of the public corporation. Harvard Business Review, 5, 61–74.
Kahneman, D., & Tversky, A. (1984). Choices, values, and frames. American Psychologist, 39, 341–350.
Khanna, N., & Poulsen, A. B. (1995). Managers of financially distressed firms: Villains or scapegoats?
The Journal of Finance, 50(3), 919–940.
Kile, C., & Phillips, M. (2009). Using industry classification codes to sample high-technology firms:
Analysis and Recommendations. Journal of Accounting, Auditing and Finance, 24(1), 25–58.
Managerial Ability and Employee Productivity 177

Kim, T. M. (2012). Distressed company management: the crucible of leadership. Journal of Corporate
Renewal, 25(8), 24–27.
Koch, M., & McGrath, R. (1996). Improving labor productivity: Human resource management policies
do matter. Strategic Management Journal, 17(5), 335–354.
Kor, Y. Y. (2003). Experience-based top management team competence and sustained growth.
Organization Science, 14(6), 707–719.
Kor, Y. Y., & Mahoney, J. T. (2005). How dynamics, management, and governance of resource deploy-
ments influence firm-level performance. Strategic Management Journal, 26(5), 489–496.
Kren, L. (1992). Budgetary participation and managerial performance: The impact of information and
environmental uncertainty. The Accounting Review, 67(3), 511–526.
Lado, A. A., & Wilson, M. C. (1994). Human resource systems and sustained competitive advantage:
A competence-based perspective. Academy of Management Review, 19(4), 699–727.
Libby, R., & Tan, H. T. (1994). Modeling the determinants of audit expertise. Accounting, Organizations
and Society, 19(8), 701–716.
Lichtenstein, B. M. B., & Brush, C. G. (2001). How do ‘resource bundles’ develop and change in new
ventures? A dynamic model of longitudinal exploration. Entrepreneurship Theory and Practice,
25(3), 37–58.
Lippman, S. A., & Rumelt, R. P. (2003). A bargaining perspective on resource advantage. Strategic
Management Journal, 24(11), 1069–1086.
Little, P. L., Little, B. L., & Coffee, D. (2009). The du pont model: Evaluating alternative strategies in
the retail industry. Academy of Strategic Management Journal, 8, 71–80.
Lowe, E. A. (1971). On the idea of a management control system: Integrating accounting and manage-
ment control. Journal of Management Studies, 8, 1–12.
Matˇejka, M., Merchant, K. A., & Van der Stede, W. M. (2009). Employment horizon and the choice
of performance measures: Empirical evidence from annual bonus plans of loss-making entities.
Management Science, 55(6), 890–905.
Merchant, K. A. (1985). Control in business organizations. Cambridge, MA: Ballinger.
Merchant, K. A., & Van der Stede, W. A. (2007). Management control systems: Performance measure-
ment, evaluation and incentives (2nd ed.). Harlow: Prentice Hall. ISBN 9780273708018.
Miller, D., & Shamsie, J. (1999). Strategic responses to three kinds of uncertainty: Product line simplic-
ity at the Hollywood film studios. Journal of Management, 25(1), 97–116.
Milliken, F. (1987). Three types of perceived uncertainty about the environment: State, effect, and
response uncertainty. Academy of Management Review, 12, 133–143.
Milner, J., & Pinker, E. (2001). Contingent labor contracting under demand and supply uncertainty.
Management Science, 47(8), 1046–1062.
Mollick, E. (2012). People and process, suits and innovators: Individuals and firm performance.
Strategic Management Journal, 33(9), 1001–1015.
Morrow, J. L., Sirmon, D. G., Hitt, M. A., & Holcomb, T. R. (2007). Creating value in the face of
declining performance: Firm strategies and organizational recovery. Strategic Management
Journal, 28(3), 271–283.
Otley, D. (1994). Management control in contemporary organizations: Towards a wider framework.
Management Accounting Research, 5(3–4), 289–299.
Penrose, E. T. (1959). The theory of the growth of the firm. New York, NY: John Wiley.
Petersen, M. (2009). Estimating standard errors in finance panel data sets: comparing approaches.
Review of Financial Studies, 22(1), 435–480.
Porter, M., & Kramer, M. (2006). The link between competitive advantage and corporate social respon-
sibility. Harvard Business Review, 84(12), 78–92.
Rajan, R., & Zingales, L. (1998). Power in a theory of the firm. Quarterly Journal of Economics, 113, 387–432.
Schultz, T. W. (1961). Investment in human capital. The American Economic Review, 51(1), 1–17.
Senbet, L. W., & Wang, T. Y. (2012). Corporate financial distress and bankruptcy: A survey. Foundations
and Trends in Finance, 5(4), 243–335.
Simons, R. (1995). Levers of control: How managers use innovative controls systems to drive strategic
renewal. Boston, MA: Harvard Business School Press.
Sirmon, D. G., Gove, S., & Hitt, M. A. (2008). Resource management in dyadic competitive rivalry: The
effects of resource bundling and deployment. Academy of Management Journal, 51(5), 919–935.
178 DIPANKAR GHOSH ET AL.

Snyder, N. H., & Glueck, W. F. (1982). Can environmental volatility be measured objectively? Academy
of Management Journal, 25(1), 185–192.
Spreitzer, G. M., McCall, M. W., & Mahoney, J. D. (1997). Early identification of international execu-
tive potential. Journal of Applied Psychology, 82(1), 6–29.
Stuebs, M., & Sun, L. (2010). Business reputation and labor efficiency, productivity, and cost. Journal
of Business Ethics, 96(2), 265–283.
Tosi, H., Aldag, R., & Storey, R. (1973). On the measurement of the environment: An assessment of
the Lawrence and Lorsch environmental uncertainty subscale. Administrative Science Quarterly,
18, 27–36.
Tushman, M., & Romanelli, E. (1985). Organizational evolution: A metamorphosis model of conver-
gence and reorientation. Research in Organizational Behavior, 7, 171–222.
Whitaker, R. B. (1999). The early stages of financial distress. Journal of Economics and Finance, 23(2),
123–133.
Widener, S. K. (2004). An empirical investigation of the relation between the strategic use of human
capital and the design of the management control system. Accounting, Organizations and
Society, 12, 377–399.
Wright, P. M., & Snell, S. A. (1998). Toward a unifying framework for exploring fit and flexibility in
strategic human resource management. Academy of Management Review, 23(4), 756–772.
Managerial Ability and Employee Productivity 179

APPENDIX: VARIABLE DEFINITIONS

Variable Description
LP (employee Net income before employee cost divided by the
productivity) number of employees
LE (employee efficiency Net income before employee cost divided by
component) employee cost
LC (employee cost Employee cost divided by the number of
component) employees
ln LP The natural log of LP (employee productivity)
ln LE The natural log of LE (employee efficiency
component)
ln LC The natural log of LC (employee cost component)
MASCORE Managerial ability score by P. Demerjian et al. (2012)
Lag_MASCORE MASCORE of period t − 1
CEO_media_citation The natural log of the number of press citations for a
CEO over a prior five-year period (Baik et al., 2009)
CDF The average of the CDF of ROA for each CEO-
firm-year of within industry-year rankings for the
prior three years (Rajgopal et al., 2006)
EMPPRO_ALT Employee performance ratings provided by the
MSCI’s ESG database
SIZE Natural log of total assets
MTB Market-to-book ratio [(CSHO*PRCC_F)/CEQ]
LEV Leverage ratio [(DLTT/AT)]
ROA Return on assets ratio
ASSETAGE Net property, plant, and equity (PPE) divided by
gross property, plant, and equipment (PPE). A
large value of ASSETAGE indicates the age of
long-term assets is young
LABORINTEN Employee cost divided by total sales
ZSCORE (financial 3.3*(net income/assets) + 0.99*(sales/assets)
distress) + 0.6*(market value of equity/liabilities) +
1.2*(working capital/assets) + 1.4*(retained
earnings/assets)
EU (environmental A five-year rolling average of sales volatility
uncertainty)
ASSETS Total net assets
180 DIPANKAR GHOSH ET AL.

Variable Description
SALE Total net sales
NI Net income
XLR Total employee cost
EMP The number of employees
CAPINTEN The ratio of capital expenditure (CAPX) over the
total assets (AT)