The Efficient Use of Information Technology:
An Industry-Level Analysis
Matt Wimble
Michigan State University
wimble@bus.msu.edu
Vallabh Sambamurthy
Michigan State University
sambamurthy@bus.msu.edu
Roger Calantone
Michigan State University
rogercal@bus.msu.edu
Abstract
Despite the salience of information technology investments in many firms,
industry and firm-specific variations exist in the efficient use of information
technology. Recent evidence suggests that industries vary in terms of how they
use information technology (IT) and that the business benefits that firms receive
from IT are strongly influenced by industry factors. This research examines some
of factors that influence variations in the efficient use of IT resources across
industries. Using COMPUSTAT data from the years 1998-2004 and data from the
Bureau of Economic Analysis (BEA) and Bureau of Labor Statistics (BLS), we
apply the Data Envelope Analysis (DEA) to trace industry-level efficiency of IT
capital. Subsequently, our analysis of covariates reveals that the competitive
intensity of the industry, industry growth, industry outsourcing intensity, and
industry capital intensity significantly influence the extent to which industries are
efficient in their use of IT. We find industry growth rate has significant
interactions with both industry concentration and capital intensity. More
interestingly, we find contrasting results about the relationships between these
factors and IT use efficiency across the services and manufacturing sectors.
Introduction
Recent evidence has suggests that industries vary in terms of how they use information
technology (IT) and the business impacts that firms receive from IT are strongly
influenced by these industry factors. Investigation of the economic impacts of IT
spending has primarily looked at the firm-level of analysis, this paper looks at the
impacts of information technology at the industry-level in order to investigate what
industry factors influence differences in economic outcome resulting from IT spending.
Further more this research note looks at IT form an efficiency lens, which is markedly
different from the central-tendency measures commonly used in economic analysis of IT
expenditure. The paper performs two analyses. First, the paper identifies key industrylevel factors the impact the efficient use of IT. Second, using a exploratory approach the
paper demonstrates how these factors are have very different effects in manufacturing
industries compared with service industries. The findings of this note are: 1) industry
concentration, outsourcing intensity, and capital intensity impact industry-level IT
efficiency, 2) industry growth rate moderates the impact of industry concentration, 3)
capital intensity moderates the impact of industry growth and 4) the impact of these
factors vary significantly between manufacturing and services.
1
Concept Development
Anecdotal evidence and practitioner studies indicate that industries differ in the extent to
which they use information technology as well as the effectiveness with which they
leverage IT functionalities and capabilities (Farrell, 2003). This study investigates why
efficiency in the use of IT resources varies across industries. Following the logic of the
research question, this review will first provide an overview of existing literature on
empirical industry-level IT economic studies and then provide an overview literature of
possible industry-level explanatory variables. Before beginning discussions of industrylevel studies it is important to discuss relevant data issues. In 1997 the U.S. census
replaces the 1987 standard Standard Industrial Classification (SIC) with the North
American Industry Classification System (NAICS) system which provided greater
differentiation for newer industries and the change of definitions lie at the heart of the
inability of researchers (Baily and Gordon, 1988; Dedrick et. al., 2003) to make definitive
industry-level statements about the impact of IT spending. For example as late as 1997
researchers were unable to detect noticeable effects from IT investments and were limited
to study only the manufacturing sector of the economy (Morrison, 1997).
Our study focuses upon the impact of IT spending upon efficiency at the industry-level.
It is necessary to frame the focus of our study before beginning an overview of existing
empirical examinations of the industry-level economic studies of IT investment.
Efficiency measures (Charnes et. al, 1994) such as Data Envelopment Analysis (DEA)
and Stochastic Frontier Analysis (SFA) provide measures relative to the efficiency
bounds are and of central tendencies. To date most industry-level research has been
central-tendency research, as that is type of research that is of most interest to
economists, but we are attempting to provide a study that will have prescriptive value to
mangers. As a result our research will have a different focus, but the central-tendency
research can be used to inform both what the approaches and data sources of the industrylevel analysis of the economic impacts of IT spending are. The most recent literature
review (Dedrick, et. al., 2003) provided a point of reference for the author and provides
an excellent overview empirical research on the economics of IT investment at the
process, firm, industry, and country-level. Empirical studies of industry-level economic
impacts of IT spending can categorized as focusing on IT-producing industries and ITconsuming industries. Also studies vary widely in the degree to which they study industry
level effects, as many are using the IT consuming or producing as an input factor to
explain productivity effects at the macroeconomic country-level and do not provide
industry-by-industry analysis. This review will focus first on industry-level studies where
the primary focus is the industry-level effects, which will be referred to as industrycentric studies. Second the review will include macro-centric studies on occasion when it
can inform this research as to possible data sources and relevant variables.
The first study (Gordon, 2000) to address industry-level impacts of IT looked at the
difference between labor productivity between industries and found that labor
productivity growth was coming in large part from the IT producing industries. The
Council of Economic Advisors (2001) was the first to report differences in labor
productivity between IT-intensive and non IT-intensive industries but provide detailed
2
analysis beyond country-level aggregates over multiple years. Bailey and Lawrence
(2001) were the first to show labor productivity growth based upon intensity of IT
consumption, but the paper did not present detailed regression results. Stiroh (2002)
produced the first industry-centric study to show industry-by-industry level effects of IT
consumption with several measures of intensity of IT consumption and showed gains
beginning in 1995 for both IT-consuming and IT-producing industries. Within
information systems (IS) literature the only published study (Han et. al., 2005) to use
industry-level data looked at the impact of IT-services industry as a proxy for outsourcing
and its impact upon productivity via a Cobb-Douglas production function using BEA
data. One study (Chang and Gurbaxani, 2005) has looked at industry-level efficiency
using a SFA approach and found that firms in more competitive markets use IT more
efficiently. A summary of the IT-consuming studies that focus at industry level are
presented in table 1.
Table 1. IT-Consuming/Industry-centric studies
Cite
Data
Source
DVs
IVs
Model
Findings
Stiroh, 2002,
AER, v92
no5, pp15591576
BEA GPO &
capital stock
data, 2-digit
Ln of output per
employee
>1995 dummy
(D), IT Capital
(C), C*D, 4
measure of
intensity
ln Ai ,t    D  IT  IT * D   i ,tInvest in late 1980s
Baily and
Lawrence,
2001, AERAEA Papers
BEA
Labor
productivity
growth
IT Intensity
No regression presented-short
conference paper.
Take-off after 1995
Han,
Kauffman,
and Nault,
CIST 2005
BEA I/O,
FA, BLS
Output
Non-IT Capital,
non-IT Labor, IT
capital, non-IT
services, IT
services
Y  AK  L M  Z K Z L log/log
Outsource positive for
high IT, no impact for
low IT
Chang and
Gurbaxani,
2005,
working
paper
CII and
Compustat
Mark-up ratio
gains by 1995
K, L, IT, IT/L,
HHI
regression
Stochastic frontier
Firms use IT more
efficiently in more
competitive markets,
high market power
makes less efficient
Beyond the industry-centric studies there are a series of papers that use industry-level IT
investment as an input for broader analysis of macroeconomic phenomena that could
inform this work in regards to potential findings, possible data sources and relevant
variables. Using BEA data from 1973-1991 Stiroh (1998) found little impact on
productivity in IT-using industries, but found positive impacts from IT-producing
industries. Stiroh (1998) used IT capital as the measure of IT usage, but did not include a
service component because the data was not available based upon the SIC coding
scheme. IT contributions to industry-level were used (Basu et. al., 2001) to study
aggregate gross output from 1987-1999 and found IT consuming to have positive effects
after 1995. Another series of papers (Basu et. al. 2003; Van Ark and Inklaar, 2005)
compared macro-level productivity effects from IT between countries using IT related
industry effects as input factors to the overall productivity functions. Comparisons
between U.S. and U.K. (Basu et. al. 2003) were found to be feasible. Direct industry-to-
3
industry comparisons were not feasible across the entire E.U. IT-producing sectors could
be examined, but IT-consuming effects could only be assumed indirectly. A summary of
the macro-centric IT studies are shown in table 2.
Table 2. Macro-centric IT Industry-level studies
Cite
Data Source
DVs
IVs
Setting/Context
Findings
Stiroh, 1998,
Economic Inquiry,
pp175-191
BEA
Gross Output,
value Added
K, L, ITK
U.S. Country, Industry
1973-1991
IT
Producing
contribution, IT using
little.
Basu, Fernald, and
Shapiro, 2001,
NBER working
paper 8359
BEA,BLS, BEA
Capital
Gross Output
N/A
Industry US 1987-1999
Productivity
in
durable
and
IT
producing, but not in
others 1995
Basu, Fernald,
Oulton, and
Srinivasan, 2003
NBER Working
Paper 10010
BEA, BLS, U.K
National Office
of Statistics
TFP
K, L, ITK, IT
Services, HW, SF,
Comm
U.S. U.K. Industry 19801995
U.S.-IT capital and
services are positive,
UK-services
are
positive, capital is
negative
Van Ark and
Inklaar, 2005,
working GD-79
BEA
STAN
Gross Output
K, L, ITK
Country & Industry U.S
EU-15
1987-2004
Industry
1995-2003:
France,
Germany,
Netherlands, U.K.
No
positive
in
Europe, positive in
US
OECD
Key points are: a) IT industry-level studies have traditionally focused upon measures of
central tendency that are of great interest to economists, but tern to be less prescriptive to
managers and b) few IT industry-level studies have looked at the industry-level effects of
consuming IT at the industry-level, but rather have looked at the impact of IT-related
industries on the macroeconomy and are thus of less interest to information systems
researchers.
A search for variables that have been used in past studies as industry-level constructs
was conducted. Our search progressed by 1) looking for IT-related industry-level
constructs, 2) finding industry-level constructs used in economic studies, and 3) industrylevel constructs used in other literature. IT-related measures consist of measures of ITintensity and IT-outsourcing within an industry. IT-related measures are presented in
table 3. All of the observed constructs can be derived using BEA capital investment and
input-output tables.
Table 3. Industry related covariates
Construct
Author
Year
Data source
Journal
IT
Capital
Intensity
(difference in difference,
absolute, & relative )
McGuckin and
Stiroh
2001
BEA
J.
of
Transfer
IT Capital Intensity (invest
rate)
Stiroh
Dec
2002
BEA
AER
regression
Herfindahl-Hirschman
Index (HHI)
Hirschman
1964
BLS/BEA/COMPUSTAT
AER
N/A
4
Method
Technology
Regression
Next, we examined existing industry-level economic variables used in industry-level
studies. Measurement of industry concentration via the Herfindahl-Hirschman Index
(HHI) occurs in numerous studies and is calculated by summing the squared marketshare
of the firms in a market.
Theoretic Development
Industrial Organizational Factors
Industrial concentration measures the number and marketshare of major competitors in a
given industry and has been the most widely studied factor in industrial organization
literature. Increased concentration has been shown to correlate with both increased
profitability and increased cost efficiency (Peltzman, 1971; Azzam, 1997). In line with
conventional wisdom, increased profitability is believed to be the result of increased
market power leading to super marginal-cost pricing. The increase in concentration
results in increased ability to control prices and increased bargaining power with
suppliers. Gains in IT efficiency should arise because of two factors, substitution effects
coupled and optimal scale economies. Prior research on industrial organization has shown
industry concentration to be a function optimal plant sizes (Weiss, 1963; Curry and
George, 1983). First, firms can expand to a certain point, after which diminishing returns
to scale limit the size of firms in a given industry. Second, IT has been shown to be a
substitute for other factors of production (Dewan and Min, 1997). As a result of this
substitution the optimal size in a given industry should increase, while the level of IT
capital remains the same. Because IT exhibits increasing returns to scale the substitution
of IT for other production factors this should lead to increased IT capital efficiency.
While one could argue the because firms in a more concentrated industry have greater
control over inputs and prices they would be less concerned with efficiency, research on
the impacts of increased concentration have consistently shown positive efficiency effects
when markets become more concentrated. Given the consistent results from industrial
organization literature coupled with the unique nature of the increasing returns to IT
capital assets we argue that:
Hypothesis 1: Increasing industry concentration is positively associated
with IT capital efficiency.
Industrial organization literature has shown that in growing industries it is easier for new
firms to enter a given market. New market entrants are more likely to use newer
technology, because they are often not subject to switching costs that arise from
upgrading technology. Despite the advantages of new market entrants, one could argue
that in growing industries firms have greater managerial slack and can be with overly
concerned with capital preservation resulting in less than optimal levels of IT investment.
Also, once could argue that firms in growing industries are chasing revenue and are often
not overly concerned with efficiency.
5
Despite the potential reasons why growth might lower efficiency, we argue that the
advantages to new firms in growth industries outweigh the potentially negative factors.
Research on the role of new technology in improving efficiency has also shown growth to
be a key factor in explaining whether a technology will result efficiency gains (David,
1990; Akeson and Kehoe, 2007). In growing industries new capacity is often built by
new firms using modern technology, but in established slow-growth industries
investment is often in inferior technologies where the firms have an existing investment.
For example, in a growth industry firms are likely to invest in the latest computing
architecture, but in more established industries firms are more likely to make investment
in legacy architecture that they already have significant investment in. This leads to the
following:
Hypothesis 2: Increased industry growth rate is positively associated with
IT capital efficiency.
According to Ghemawat and Nalebuff (1984) increased efficiency from increase
concentration is related to industry expansion. Gains in efficiency attributable to growth
are likely result from newer and smaller firms with a superior technological advantage
lowering average cost and thus expanding the overall market. As discussed above there
are potential reasons why increases in both growth and concentration could result in
lower efficiency. In markets that experience both increases in concentration and increases
in growth, the growth is likely result from existing firms, which are often subject to
substantial switching costs to change technologies. Also these not are likely to have less
concern for efficiency and concentrate on expanding to meet the increased demand. This
leads to the following:
Hypothesis 3: Increased industry growth rate will negatively moderate the
impact of increased industry concentration on IT capital efficiency.
Capital-intensive industries have several characteristics that should lower IT efficiency.
First, increased capital intensity raises the barrier to entry for new firms, which in turn
results in lower competition and diminished efficiency (Capon, et. al., 1990; Bharadwaj,
et. al. , 1999). Also, increased capital investment is also likely to take resources away
from complementary investments that are necessary with IT investments (Bharadwaj, et.
al. , 1999). However, because IT has been shown to be a substitute for ordinary capital
IT efficiency could be grater in capital intensive industries because there is a greater
potential for gain though substitution. Despite the fact that IT has been shown to be a net
substitute for ordinary capital in aggregate at the firm level, this does not mean that IT
can always substitute for ordinary capital. It could very well be the case that in capital
intensive industries, like heavy manufacturing or metal production, that the substitution
of IT is quite limited. In capital-intensive industries the capital often takes the form, such
as a stamping press and smelters, such that there is likely no suitable substitute. As a
result in capital-intensive industries the gains from substitution are likely to be
6
outweighed by the cost to efficiency of lower competition and underinvestment due to
resource demands of ordinary capital. This leads to the following:
Hypothesis 4: Increasing capital intensity is negatively associated with IT
capital efficiency.
As discussed above, increased capital intensity raises the barrier to entry for new firms
(Capon, et. al., 1990; Bharadwaj, et. al. , 1999). New firm entry has shown to be a way
in which industry efficiency improves and is likely to be even more pronounced for IT
efficiency due to switching cost issues. Since increased capital intensity makes it more
difficult for new firms to enter the market and that new firm entry is key to efficiency
gains resulting from industry growth, this leads to the following:
Hypothesis 5: Increasing capital intensity will negatively moderate the
impact of growth on IT capital efficiency.
Transaction Cost Factors
Buyer/supplier relations are a central subject of research in transaction cost economics
(TCE) and have a very long history of both empirical and theoretic work (Shelanski and
Klein, 1995). IT has been shown to reduce external coordination costs by improving the
monitoring of suppliers (Bakos and Brynjolfsson, 1993). IT has also been shown to
reduce agency costs by reducing the cost of monitoring employees, which could result in
IT efficiency being greater in industries where a greater share of production is
internalized. Prior research has shown that while IT does reduce internal coordination
costs, it reduces external coordination costs to a greater degree (Brynjolfsson, et. al.,
1994). Consistent with TCE, by reducing external coordination cost through IT, firms are
more effectively able to manage suppliers and externalize inefficient internal operations
(Malone, Yates, and Benjamin, 1987). Given that prior research has shown that a major
benefit of IT is that it increases the ability to externalize, or outsource, inefficient
operations leading to increased efficiency , this leads to the following:
Hypothesis 6: Increasing outsourcing intensity is positively associated
with IT capital efficiency.
Comparison of Services and Manufacturing
Finally, we explore differences in IT efficiency between manufacturing and services.
Services differ from manufacturing because of the nature of production in a service
context is inherently different from production in a manufacturing context. Services
exhibit the characteristics of intangibility, inseparability, and heterogeneity. Intangibility
refers to the idea that services cannot be inventoried, are not readily measured, and they
7
do not even consume physical space (Shostack, 1977). Inseparability refers to the idea
that the consumption of a service and the production of a service often occur
simultaneously (Carmen and Langeard, 1980). Service production is often inseparable
from consumption to such a degree that the consumer rises to the level of co-production
(Parasuraman, et. al., 1985). Heterogeneity refers to the idea that services often vary from
day to day and customer to customer (Parasuraman, et. al., 1985). Services and
manufacturing are different in that in a service context the customer supplies key inputs
to the production process (Brown, et. al. 2002). Co-production of output that is common
in services necessitates a high degree of cooperation between consumer and producer. In
service industries the production process is highly contingent upon the specific
interactions of consumers and producers, which implies far greater uncertainty a priori in
the sequence of events necessary for production of services. As a result high degree of
uncertainty results from the co-production found in services (Argote, 1982; Jones, 1987).
The heterogeneity inherent in service processes manifests as variety that can be seen as a
sign of the flexibility that is necessary for high quality (Feldman, 2000). In a
manufacturing environment, in contrast, variation in the sequence of tasks used in
production is seen as indicative of poor quality (Oakland, 1996). Empirical work on task
sequencing has observed a high degree of variety in service settings (Pentland, 2003).
Previous studies have shown processes to be a potential source of flexibility in
organizations (Feldman and Pentland, 2003). Increasingly, information processing
involves the use of workflow management systems, which are being used to define work
processes in service industries (Fletcher, et. al., 2003). The ability of a service provider to
deal with a wide variety of situations is a mark of high customer service (Zeithaml, et. al.,
1990; Cronin and Taylor, 1994) and a key factor in retaining customers in service
environments (Keaveney, S., 1995). Service workers must be capable of developing
novel solutions to the often unique situations they often face. A great deal of uncertainty
results from this uniqueness, often requiring a great deal of information processing and a
high level of IT capital (Bowen and Ford, 2002).
Productive effects from IT investment are known to arise as a result of IT capital
substituting for labor (Dewan and Min , 1997). Comparisons of manufacturing and
services have long observed that substitution of capital for labor is much easier in
manufacturing than in services (Baumol, 1967). As an example it takes a nurse today
nearly as long to change a bandage as it did a hundred years ago, but the manufacture of
most products over this time has required drastically less labor. Prior empirical work has
found that manufacturing industries have greater gains in terms of productivity form IT
than services (Dewan and Min , 1997). In summary, services differ strongly from
manufacturing in terms of intangibility, inseparability, and heterogeneity. These
differences have manifested in empirically observable differences in the effect of various
input factors, both IT and non-IT, have upon productivity. Due to these substantial
differences it is expected that the impact of industry factors on the efficient use of IT will
vary from services to manufacturing, which leads to the following proposition:
8
Proposition 1: The correlation of industry-level factors on IT capital
efficiency will vary significantly from manufacturing to services.
Data
The modeling approach used in this paper is a two-stage analysis. First, efficiency scores
are generated using DEA analysis and next we explain the differences in efficiency using
regression analysis. The explanation of the data follows the flow of the analysis. The
Data for this study was collected from several sources. The data is United States industrylevel data 3-digit NAICS granularity for the seven years from 1998-2004. The paper
includes 43 industries per year, for a total of 301 industry- years. A summary of the data
used to develop the efficiency scores is shown in shown in table 4. All dollar
denominated measures are in real-dollar terms, year 2000 dollars.
Table 4. Variables and data sources for DEA analysis
Variable
Description
Data source
Factor
ITK
IT capital stock. Sum of hardware, software, and
communication equipment in dollars.
BEA fixed asset tables
Input
ITO
IT outsourcing. Sum of industries 5415
“Computer systems design and related services”
and 514 “Information and data processing
services” in dollars
BEA input-output accounts
Input
K
All non-IT capital stock in dollars.
BEA fixed asset tables
Input
L
Labor expenditure in dollars.
BEA input-output accounts
Input
M
Sum of all non-IT outsourcing in dollar terms.
BEA input-output accounts
Input
YE
Output per employee. Y/E
BEA input-output accounts,
BLS industry employment
tables.
Output
VAE
Value added per employee VA/E. An industrylevel proxy for profitability.
BEA input-output accounts,
BLS industry employment
tables.
Output
A description of the covariates to explain the efficiency scores obtained in the DEA step
and control variables are shown in tables 5 and 6.
Table 5. Covariates used to explain efficiency scores.
Variable
Description
Data source
HHI
Herfindahl-Hirschman Index. A measure of industry
n
concentration.
HHI j   mi2, j
the industry and
industry.
i 1
mi , j
where n is the number of firms in
is the marketshare of the ith firm in the jth
9
COMPUSTAT
GROW
Growth rate for the industry for that year.
BEA input-output accounts
GROWHHI
The interaction between growth and industry concentration
COMPUSTAT
and BEA
input-output accounts
KINT
Capital intensity. K/E: non-IT capital per employee
BEA fixed asset tables, BLS
employment tables
GROWKINT
The interaction between growth and capital intensity
COMPUSTAT, BEA inputoutput accounts & BEA
fixed-asset tables
OINT
Outsource intensity. M/Y: non-IT outsourcing divided by gross
output.
BEA input-output accounts
SERV
Binary indicating a service industry NAICS 511 through 81.
BEA input-output accounts
SHHI
Herfindahl-Hirschman Index in a service industry
COMPUSTAT
SGROW
Growth rate for the industry for that year in a service industry
BEA input-output accounts
SHHIGROW
The interaction between growth and industry concentration in a
service industry
COMPUSTAT
and BEA
input-output accounts
SKINT
Capital intensity. K/E: non-IT capital per employee in a service
industry
BEA fixed asset tables, BLS
employment tables
SGROKINT
The interaction between growth and capital intensity in services
COMPUSTAT, BEA inputoutput accounts & BEA
fixed-asset tables
SOINT
Outsource intensity. M/Y: non-IT outsourcing divided by gross
output in a service industry
BEA input-output accounts
Table 6. Control variables
Variable
Description
Data source
ITKINT
IT capital intensity. ITK/E: IT capital per employee
BEA fixed asset tables, BLS
employment tables
SITKINT
IT capital intensity. ITK/E: IT capital per employee in a service
industry
BEA fixed asset tables, BLS
employment tables
YEAR
Variable from 0-7 indicating the year.
N/A
The industry concentration measures were calculated from COMPUSTAT because the
more often used industry concentration measures from the US Economic Census are only
calculated for manufacturing industries. We calculated the growth rates as one-year
estimates for two reasons. First, the NAICS data we used starts in 1997, prior to that the
industry data was calculated using the SIC scheme. Second, as discussed in the theory
section the entry is key to our argument and this is best studied using contemporaneous
growth rates (Hause and Du Rietz, 1984; Bloch, 1981). Consistent with prior literature
the outsource intensity measures used are the level of outsourcing in dollar terms relative
to the level of gross output (Feenstra and Hanson, 1996). Capital intensity measures are
measured in dollar per employee terms (Stiroh, 2002). The control for time is an integer
from 0 to 6.
10
DEA efficiency
The first step in our analysis was to obtain efficiency scores for each industry-year using
data envelopment analysis (DEA). DEA measures offer several advantages. First DEA
measures are inherently prescriptive, as opposed to the descriptive nature of centraltendency measures such as ordinary least squares regression. Second, DEA allows for the
combination of multiple inputs and multiple outputs into a single virtual input and a
single virtual output. DEA has become an increasingly important means to investigate
efficiency due the flexibility it provides. DEA provides a means to include multiple
output measures and requires no statistical assumptions be made about the data. We
calculated the efficiency scores over 61 industries for each of the seven years available on
a year-by-year basis in order to make comparisons across years and to control for changes
attributable to differences in macroeconomic conditions. The DEA efficiency scores
were calculated using the Banker/Charnes/Cooper (BCC), formulation (Banker, et. al.,
1984). The DEA analysis used was a convex hull, output-oriented, variable returns to
scale (VRS) formulation. We chose to use a VRS implementation in order to account for
the scale differences between industries. Resulting from the DEA is a set of slacks for
each input term. The resulting slacks are used to obtain an efficiency score, in percentage
terms, for the IT-related input factor of IT capital.
Covariate analysis
We performed the covariate analysis using OLS regression. Recent research has shown
OLS applied in the second stage of a DEA efficiency analysis to be a superior approach
to either a parametric approach, such as stochastic frontier, or a Tobit regression on the
second stage of a DEA score for analysis of the impact of exogenous covariates on
efficiency (Banker and Natarajan, 2007). The approach this paper takes has been
successfully applied to efficiency analysis in other contexts (Ray, 1991). After obtaining
efficiency scores for all industry/years, the service and manufacturing industries were
separated from the whole data set for analysis. We compared manufacturing industries,
those with NAICS codes in the 300s, to pure service firms. Although transportation and
wholesale/retail trade is sometimes considered a service, they were not included because
they do not meet the classic criteria of a service. Transportation and retail/wholesale trade
involve inventory, thus do not meet the intangibility criteria of a service. Also, they
typically involve a low degree of heterogeneity in production relative to pure services
such as consulting or food service. As a result, the study defines services as NAICS codes
511 through 81. Three regressions were used. The first model consists of the IT capital
efficiency score regressed against the market concentration ratio, industry growth rate, an
interaction between industry growth and industry concentration, outsource-intensity, and
capital intensity. The second model is same as the first, but also contains a binary
indicator for service industries. The third model is an exploratory model used to develop
separate estimates of effects of each of the factors in the main model of manufacturing
and services. Endogeneity was controlled for using IT capital intensity and time was
controlled for using an integer representing the year. The equations are used are as
follows:
11
I)
ITKEFF   0   1 HHI   2 GROW   3 GROWHHI   4 KINT
  5 GROWKINT   6 OINT   7 ITKINT   8 YEAR  
II)
ITKEFF   0   1 HHI   2 GROW   3 GROWHHI   4 KINT
  5 GROWKINT   6 OINT   7 ITKINT   8 SERV   9 YEAR  
III)
ITKEFFi , t   0   1 HHI   2GROW   3GROWHHI   4 KINT   5GROWKINT
  6OINT   7 ITKINT   8 SERV   9 SHHI   10 SGROW   11SHHIGRO
  12 SKINT   12 SGROKINT   13STOINT   14 SITKINT   15YEAR  
Results
The purpose of this study is to investigate why efficiency in the use of IT resources varies
across industries. Due to the cross-sectional nature of the data we checked for
heteroskedasticity using the White Heteroskedasticity Test with cross-terms on all
regressions and corrected for it using White Heteroskedasticity-Consistent Standard
Errors (WHCSE) where indicated. Finally the IT efficiency scores were highly leftskewed, which was corrected using an inverse-log transformation. All regressions were
performed in STATA. Regression results are shown in table 7.
Table 7. Regression Results
Industry Concentration+
Industry Growth Rate+
Growth/Concentration Interaction+
Capital Intensity+
Growth/Capital Interaction+
Outsourcing Intensity+
Services Binary
Industry Concentration in Serv.
Industry Growth Rate in Serv.
G/HHI Interaction in Serv.
Capital Intensity in Serv.
G/K Interaction in Serv.
Outsourcing Intensity in Serv.
IT Capital Intensity+
IT Capital Intensity in Serv.
Year
I: Base Model
II: Model with Services
Binary
III: Model with Separate
Coefficients for Services
and Manufacturing
0.17 (.11)***
0.10 (3178.03)*
-.13 (1.24)**
-.18 (20.69)**
-.10 (197.51)*
.15 (1938.47)**
N/A
N/A
N/A
N/A
N/A
N/A
N/A
-.10 (43.52)
N/A
.01 (122.36)
0.17 (0.11)***
0.10 (3171.3)*
-0.13 (1.25)**
-0.18 (20.67)**
-0.10 (197.85)*
0.16 (2073.24)**
0.01 (536.38)
N/A
N/A
N/A
N/A
N/A
N/A
-0.10 (44.81)
N/A
0.02 (122.79)
-0.33 (0.31)**
-0.34 (7547.04)**
0.49 (4.16)**
0.98 (36.04)***
0 (281.37)
-0.47 (4550.01)***
-1.71 (2879.35)***
0.48 (0.33)***
0.46 (8096.67)***
-0.6 (4.25)***
-1.22 (38.27)***
-0.01 (350.91)
1.14 (5037.99)***
-7.63 (494.41)***
7.8 (496.41)***
-0.04 (104.63)
F
5.41
4.82
Prob > F
0.00
0.00
R-squared
0.12
0.12
Root MSE
4113.80
4120.80
Note: + in model 3 these represent the values in manufacturing.
*,**,*** indicate significance at p=.10, .05, .01 repectively
42.62
0.00
0.39
3462.40
The results from the base model indicate increased industry concentration and
outsourcing intensity are positively correlated with increased IT efficiency. Industry
12
growth rate was not a significant covariate, but was positively correlated with IT capital
efficiency. The interaction between industry concentration and industry growth rate was
significant and negatively correlated. Results from base model also indicate that neither
of the control factors, IT capital intensity and time, did not have significant effects.
Overall model fit was good. Model two results with a binary variable for service
industries do not indicate significant differences in IT capital efficiency between services
and manufacturing and findings were consistent with the base model. The summary of
findings is shown in table 8 and 9.
Table 8. Summary of findings for main models I & II
Hypothesis
Findings
Support?
H1: Increasing industry concentration is positively associated
with IT capital efficiency.
Significant at 1%
Yes
H2: Increasing industry growth is positively associated with
IT capital efficiency.
Significant at 10%
Partial
H3: Increased industry growth rate will negatively moderate
the impact of increased industry concentration on IT capital
efficiency.
Significant at 5%
Yes
H4: Increasing capital intensity is negatively associated with
IT capital efficiency.
Significant at 5%
Yes
H5: Increasing capital intensity will negatively moderate the
impact of growth on IT capital efficiency.
Significant at 10%
Partial
H6: Increasing outsourcing intensity is positively associated
with IT capital efficiency.
Significant at 5%
Yes
Table 9. Summary of findings for proposition 1/model 3
Covariate
Manufacturing
Services
Industry Concentration (HHI)
Negative, significant at 5%
Positive, significant at 1%
Yes
Industry Growth (GROW)
Negative, significant at 5%
Positive, significant at 1%
Yes
Growth interaction with
Concentration (GROWHHI)
Positive, significant at 5%
Negative, significant at 1%
Yes
Capital Intensity (KINT)
Positive, significant at 1%
Negative, significant at 1%
Yes
Capital Concentration
interaction with Growth
(GROWKINT)
Effect not significant
Effect not significant
No
Outsource Intensity (OINT)
Negative, significant at 1%
Positive, significant at 1%
Yes
IT Capital Intensity (ITKINT)
Negative, significant at 1%
Positive, significant at 1%
Yes
13
Support?
Limitations
This study is not without limitations. There are four primary limitations to this study. The
first limitation is that since the study included service industries the industry
concentration was induced from COMPUSTAT, rather than from the Economic Census.
Secondly, this study is of a rather limited time frame. While IT assets data is available
through the industry accounts back until the 1960s, two factors make use of this data
inappropriate. The industry accounts were redefined from SIC to NAICS in 1997, which
can bias comparisons between time periods. Prior to 1998 the make-use industry-level
accounts were not available, thus important factors such as IT and non-IT related
outsourcing could not be included in the study. Thirdly, this study looks at IT capital, but
does not provide a more detailed breakdown of IT assets. Finally, the study does not
partial out effects from IT labor separate from overall labor expenditure.
Implications for Research
Our research has provided researchers with a better understanding of what industry-level
factors impact the efficient use of IT. The research filled an important literature gaps in
several ways. First, this study examined using a frontier lens rather than a centraltendency lens. The frontier-based lens is useful in that it is both prescriptive and can
accommodate situations where multiple objectives are feasible. Second, this study looked
at IT use from an industry-level. Despite the much literature to suggests that firms often
base IT investment decisions upon within-industry benchmarking and that industry
factors play a critical role in how IT is used, little was know about what these industrylevel factors were. Finally, this study provided empirical evidence that services and
manufacturing vary considerable in how industry-factors relate to the effective use of IT.
Implications for Managers
Recent evidence from practice suggest that companies often benchmark IT practices
using within-industry comparisons and that cross-industry comparisons are much more
useful (Cullen, 2007). This study outlined industry-level forces that impact how
effectively IT is used in a given industry and that can help managers understand how IT
use varies across industries due to these factors. This study has four main findings for
managers. First, industry concentration, industry growth, the outsourcing-intensity of the
industry, and the capital intensity of the industry all critically impact how effectively IT
is used in a given industry. Second, industry growth and industry concentration have
important interactions that also impact how effectively IT is used in a given industry.
Third, the capital intensity of an industry reduces the benefits of higher industry growth.
Finally, manufacturing and services are influenced in radically different ways by these
industry forces.
Future Research
Areas of future research include performing efficiency analysis at a more granular level
such as firm-level, performing a more longitudinal analysis such as a Malmquist , and
14
using the disaggregated measures of IT capital provided in the BEA industry accounts to
discover differential effects from different types of IT capital. Also, further investigation
in terms of both theory and empirical analysis is needed to explain as to why
manufacturing and services seem to vary so drastically in terms of effects from IT.
Conclusion
This research explored several under investigated aspects of the impacts of information
technology spending. First, this paper examined the efficient use of IT, most prior studies
use central-tendency measure to examine the impacts of IT on average. Second, this
paper identified several industry-level factors that impact the efficient of IT. Finally, this
paper disaggregated industries and showed that industry concentration has different
effects on services compared with manufacturing. The paper represents a significant step
forward on those fronts. The paper performs two analyses. First, the paper identifies key
industry-level factors the impact the efficient use of IT. Second, using a exploratory
approach the paper demonstrates how these factors are have very different effects in
manufacturing industries compared with service industries. The findings of this note are:
1) industry concentration, growth, outsourcing intensity, and capital intensity impact
industry-level IT efficiency, 2) industry growth rate moderates the impact of industry
concentration, 3) capital intensity moderates the impact of growth and 4) the impact of
these factors vary significantly between manufacturing and services.
15
Works Cited
1) Atkeson, A. and Kehoe, P., 2007, “Modeling the Transition to a New Economy:
Lessons from Two Technological Revolutions”, American Economic Review,
v97(1), pp.64-88
2) Argote, L., 1982, “Input Uncertainty and Organizational Coordination in Hospital
Emergency Units, Administrative Science Quarterly, v. 27, pp. 420-434
3) Azzam, A., 1997, “Measuring Market Power and Cost-Efficiency Effects of
Industrial Concentration”, The Journal of Industrial Economics, v45(4), pp. 377386
4) Baily, M. and Gordon, R., 1988, “The Productivity Slowdown, Measurement
Issues, and the Explosion of Computer Power”, BPEA, 2, pp. 347-420.
5) Baily and Lawrence, 2001, “Do We Have a New E-conomy?”, American
Economic Review -AEA Papers, v91 n2, pp308-312
6) Bakos, Y. and Brynjolfsson, E., 1993,"From Vendors to Partners: Information
Technology and Incomplete Contracts in Buyer-Supplier Relationships," Journal
of Organizational Computing, v3(3)
7) Banker, R., Charnes, A. and Cooper, W., 1984, “Some models for estimating
technical and scales efficiencies in data envelopment analysis”, Management
Science, v. 30(9), pp. 513-21.
8) Banker, R. and Natarajan, R., 2007, “Evaluating Contextual Variables Affecting
Productivity Using Data Envelopment Analysis”, forthcoming in Operations
Research
9) Basu, Fernald, and Shapiro, 2001, NBER working paper 8359
10) Basu, Fernald, Oulton, and Srinivasan, 2003, “The case of the Missing
Productivity Growth: Or Does Information Technology Explain Why Productivity
Accelerated in the United States but Not the United Kingdom?”, NBER Working
Paper 10010
11) Baumol, W., 1967, “Macroeconomics of Unbalanced Growth: The Anatomy of
Urban Crisis”, The American Economic Review, v 57(3), pp. 415-426
12) Baumol, Panzar, and Willig, 1982, Contestable Markets and the Theory of
Industry, Harcourt Brace, San Diego.
13) Bharadwaj, A., Bharadwaj, S., and Konsynski, B., 1999, “Information
Technology Effects on Firm Performance as Measured by Tobin’s q”,
Management Science, v45(6), pp. 1008-1024
14) Bloch, H., 1981, “Determinants of the Variance in Market Shares in Canadian
Manufacturing”, The Journal of Industrial Economics, v29(4), pp 385-393
15) Bosworth and Triplett, “Services Productivity in the United States: Griliches’
Services Volume Revisited”, CRIW Conference, 2003
16
16) Bowen, J., and Ford, R., 2002, “Managing Service Organizations: Does Having a
“Thing” Make a Difference?”, Journal of Management, 28(3), pp. 447–469
17) Brown, S., Ostrom, A., Bettencourt, L., and Roundtree, R., “Client Co-Production
in Knowledge-Intensive Business Services”, California Management Review,
Summer 2002.
18) Brynjolfsson, E., Malone, T., Gurbaxani, V., and Kambil, A., 1994, “Does
Information Technology Lead to Smaller Firms?”, Management Science, v40(12),
pp. 1628-1644
19) Carmen, J. and Langeard, E., 1980, “Growth Strategies of Service Firms”,
Strategic Management Journal, v1(1), pp. 7-22
20) Chang and Gurbaxani, 2005, “An Empirical Investigation of IT Returns: the Role
of IT and Competition as Determinants of Efficiency”, working paper
21) Charnes, A., Cooper, W., and Rhodes, E., 1978, “Measuring the efficiency of
decision making units”, European Journal of Operational Research, v.43(2),
pp.429-44.
22) Charnes, A., Cooper, W., Lewin, A., and Seiford, L., 1994, Data Envelopment
Analysis: Theory, Methodology, and Application, Kluwer Academic Press,
Boston, Ma
23) Conner, J. and Peterson, E., 1992, “Market-Structure Determinants of NationalPrivate Label Price Differences of Manufactured Food Products, The Journal of
Industrial Economics, v40 no2, pp157-171
24) Council of Economic Advisors, 2001, The Economics of the President. U.S.
Government Printing Office, Washington D.C.
25) Cullen, A., 2007, “Comparing Your IT Shop To Others: Benchmark The Right
Way To Gain Useful Insights”, CIO Roadmap, Forrester Research, Cambridge,
MA, January 10, 2007
26) Curry, B. and George, K., 1983, “Industrial Concentration: A Survey”, The
Journal of Industrial Economics, v31(3), pp.203-255
27) David, P., 1990, “The Dynamo and the Computer: An Historical Perspective on
the Modern Productivity Paradox.”, American Economic Review, v80(2), pp.35561
28) Dedrick, Gurbaxani, and Kraemer, 2003, “Information Technology and Economic
Performance: A Critical Review of the Empirical Evidence”, ACM Computing
Surveys, V35, no 1, pp1-28
29) Dewan, S. and Min, C.,1997, “The Substitution of Information Technology for
Other Factors of Production: A Firm Level Analysis”, Management Science, v
43(12), pp. 1660-1675
30) Färe, R., 1986, “Addition and Efficiency”, Quarterly Journal of Economics,
v101(4), pp. 861-866.
17
31) Feenstra, R. and Hanson, G., 1996, “Globalization, Outsourcing, and Wage
Inequality”, American Economic Review, v86(2), pp. 240-245
32) Feldman, M.S., 2000, “Organizational Routines as a Source of Continuous
Change”, Organization Science, v. 11, pp.611-629
33) Feldman, M. and Pentland, B., 2003, “Reconceptualizing Organizational Routines
as a Source of Flexibility and Change”, Administrative Science Quarterly, v.
48(1), pp. 94-118.
34) Figenbaum, A. and Thomas, H., 1986, “Dynamic and Risk Measurement
Perspectives on Bowman's Risk-Return Paradox for Strategic Management: An
Empirical Study”, Strategic Management Journal, Vol. 7, No. 5, pp. 395-407
35) Filardo, A., 1997,“Cyclical Implications of the Declining Manufacturing
Employment Share.", Economic Review Federal Reserve Bank of Kansas City,
82(2), pp. 63-87.
36) Fletcher, A., Brahm, M., and Pargmann, H., Workflow Management with SAP
Webflow, Springer-Verlag, New York.
37) Ghemawat, P. and Nalebuff, B., 1990, “Declining Markets: The Devolution of
Declining Industries”, Quarterly Journal of Economics, v. 105, pp167-86.
38) Gordon, R., 2000, “Does the “new economy” measure up to the great inventions
of the past?”, Journal of Economic Perspectives, v. 14, no. 4, pp. 49-76
39) Griliches, Z., 1969, “Capital-skill Complementarity”, Review of Economics and
Statistics, 51(4), 465–468.
40) Gupta, M. and Huefner, R., 1972, “A Cluster Analysis of Financial Ratios and
Industry Characteristics”, Journal of Accounting Research, v10 n1, pp77-95
41) Gurbaxani, V. and Wang, S., 1991,"The impact of Information Systems on
Organizations and Markets," Communications of the ACM, v34(1), pp. 58-73.
42) Han, Kauffman, and Nault, “Does IT Outsourcing Payoff? Evidence from
Industry-Level Data”, CIST 2005
43) Hause, J. and Du Rietz, G., “Entry, Industry Growth, and the Microdynamics of
Industry Supply”, Journal of Political Economy, v92(4), pp 733-757
44) Hirschman, A., 1964, “The Paternity of an Index”, The American Economic
Review, Vol. 54, No. 5, p 761.
45) Jones, G., 1987, “Organization-Client Transactions and Organizational
Governance Structures”, Academy of Management Review, v. 30, pp. 197-219.
46) Keaveney, S., 1995, “Customer Switching Behavior in Service Industries: An
Exploratory Study”, Journal of Marketing, v. 59(2), pp. 71-82.
47) Malone, T., Yates, J., and Benjamin, R., 1987,“Electronic Markets and Electronic
Hierarchies,” Communications of the ACM, v30(6), pp. 484-497.
48) McGuckin, R. and Stiroh, K., 2001, “Do Computers Make Output Harder to
Measure?”, Journal of Technology Transfer, v26 n4, pp. 295-321
18
49) Morrison, C., 1997, “Assessing the Productivity of Information Technology
Equipment in U.S. Manufacturing Industries”, The Review of Economics and
Statistics, v79, n3, pp. 471-481
50) Oakland, J.S., 1996, Statistical Process Control, Butterworth-Heinemann,
Woburn, MA.
51) Parasuraman, A., Zeithaml, V., and Berry, L., 1985, “A Conceptual Model of
Service Quality and Its Implications for Future Research”, Journal of Marketing,
v. 49(3), pp. 41-50.
52) Peltzman, S., 1977, “The Gains and Losses From Industrial Concentration”, The
Journal of Law and Economics, v20(2), pp. 229-263
53) Pentland,B., 2003, “Sequential Variety in Work Processes”, Organization
Science, 14(5), pp. 528-540
54) Ray, S., 1991, “Resource-use Efficiency in Public Schools: A Study of
Connecticut Data”, Management Science, v37(12), pp.1620-1628
55) Shelanski, H. and Klein, P, 1995, “Empirical Research in Transaction Cost
Economics: A Review and Assesment”, The Journal of Law, Economics, and
Organization, v11(2), pp. 335-361
56) Shostack, G., 1977, “Breaking Free from Product Marketing”, Journal of
Marketing, v41(2), pp. 73-80
57) Simon, H., 1973,"Applying Information Technology to Organization Design,"
Public Administration Review, v33(3), pp. 268-278.
58) Stiroh, K., 1998, “Computers, Productivity, and Input Substitution”, Economic
Inquiry, pp175-191
59) Stiroh, K., 2002, “Information Technology and the Productivity Revival: What
Do the Industry-Level Data Say?”, American Economic Review , v92 no5,
pp1559-1576
60) Van Ark and Inklaar, 2005, “Catching Up or Getting Stuck? Europe’s Troubles to
Exploit ICT’s Productivity Potential”, working paper GD-79
61) Weiss, L., 1963, “Factors Changing Concentration”, Review of Economics and
Statistics, v45(1), pp.70-77
62) Williamson, O., 1973, “Markets and Hierarchies: Some Elementary
Considerations”, American Economic Review, 63: 316–325.
63) Williamson, O., 1981,”The Modern Corporation: Origins, Evolution, Attributes”,
Journal of Economic Literature, 19(4), pp.1537-1568
64) Zeithaml, V., Parasuraman, Z. and Berry, L., Delivering Quality Service:
Balancing Customer Perceptions and Expectations., New York: The Free Press,
1990.
65) Zellner, A. ,1962,"An efficient method of estimating seemingly unrelated
regression equations and tests for aggregation bias", Journal of the American
Statistical Association ,57: 348–368.
19