[an error occurred while processing this directive]
IS Productivity And Outsourcing
Policy:
A Conceptual Framework and Empirical Analysis
©
Copyright, 1995, Yogesh
Malhotra, Ph.D., @BRINT Institute, All Rights Reserved
E-Mail: malhotra@brint.com
Reference citation for this document is given below:
Malhotra, Yogesh. "IS
Productivity And Outsourcing Policy: A
Conceptual Framework and Empirical Analysis," in the Proceedings of
Inaugural Americas Conference on Information Systems (Managerial
Papers),
Association for Information Systems, Pittsburgh, Pennsylvania,
August 25-27, 1995, pp. 142-144.
This article may be printed as a paper copy for non-profit,
non-commercial, academic or educational use provided no alterations
are made and the copyright notice is maintained intact.
Any other use requires a written preapproval from
malhotra@brint.com
Published in the
Proceedings of Inaugural
Americas Conference on
Information Systems (Managerial Papers), Association for
Information Systems,
Pittsburgh, Pennsylvania, 1995
Click to View
Related Powerpoint Presentation (378KB)
IS Productivity And Outsourcing Policy:
A Conceptual Framework and Empirical Analysis
IS outsourcing policies define the criteria that organizations
utilize to decide upon the scope and
degree of reliance of their IS capabilities upon external sources.
Financial considerations and business
strategy are the two major determinants of the IS outsourcing decision.
Most controversy on outsourcing of
IS has been around the issue of increasing performance - generally by
reducing costs and improving service.
In this context, the study of the interrelationships between IS
outsourcing policy, the business and financial
strategy considerations and IS productivity, is increasingly relevant for
providing a more balanced
perspective to the ongoing debate.
This paper proposes an integrative conceptual framework for
analyzing the above relationships. A
structural equation model was created to represent the proposed
conceptual framework. The model was
operationalized by using factor analysis and stepwise regression
analyses. Since the objective of the study
was to investigate the influence of financial criteria and business
strategy variables on IS performance, the
set of companies selected as "the 100 most effective users of information
technology" by Computerworld Premier
100 survey (1994) represents an acceptable sample. Data integrity was
ensured by extracting all remaining
data from the same public sources as those used by the survey. The
study has important implications for
the new measure of IS productivity, as well as the continued emphasis on
primarily macro-level financial
measures for determining the effectiveness of IS.
1. Conceptual Framework for the Proposed Model
The proposed conceptual framework has two motives: (i) to
assess the influence of Business Strategy
and Financial Conditions on IS Outsourcing Policy, (ii) to evaluate the
relative importance of Financial
Considerations and IS Outsourcing Policy as determinants of IS
Productivity. The first objective is
characterized by:
OP = f(B,F) (1)
where the symbols denote vectors with components representing IS
Outsourcing Policies (OP), Business
Strategy (B), Financial Conditions (F), and f denotes "function of." The
other objective is represented
by:
ISP = g(F,OP) (2)
where the components of ISP represent IS productivity, and those of F and
OP are as described in Table 1,
and g denotes "function of." The two objectives are integrated in the
model delineated in Figure 1.
The endogenous and exogenous variables, and their corresponding indices,
of the conceptual model
are listed in Table 1.
Table 1. Variables Used in this Study and their Units of Measurement
____________________________________________________________________________________
Note: In the following table, the Variables are represented by the
INDICES listed under them.
Subcategories of indices presented under some variables are based on
theory and intuition.
ENDOGENOUS VARIABLES
IS Outsourcing Policy (OP)
Criteria used by organizations to decide upon the scope of their
dependence upon outside IS vendors
for meeting their needs.
ISOUT % IS Budget Spent Outside IS Department
IS Productivity (ISP)
Return from investments in IS.
IPI (Information Productivity Index)
EXOGENOUS VARIABLES
Financial Considerations (F)
Financial conditions of the company that may affect allocation of capital
and cost-cutting measures for different functions.
Business cost structure
CSSA (Cost of Goods Sold+Selling, General & Administrative Expenses)/Sales
CSTA (Cost of Goods Sold+Selling, General & Administrative
Expenses)/Total Assets
Financial leverage
LDSE Long-term Debt/Shareholders' Equity
TLSE Total Liabilities/Shareholders' Equity Business performance
REOA Return on Assets
EAPS Earnings Per Share ($)
Technology cost structure
ITGP IT Expenditure/Gross Plant, Property & Equipment
ITNP IT Expenditure/Net Plant, Property & Equipment Return on Technology
NIIT Net Income/IT Expenditure
SAIT Sales/IT Expenditure
Business Strategy (B)
Strategic significance of the IS function in the company.
Relative Importance of IS
ISBDGT IS Budget as % of Revenue
BDGTCHG Growth in IS Budget (% over last 5 years)
ISPEMP Total Number of IS Staff/Total Number of Employees
Strategic Position of IS within firm
TOCEO Categorical variable to indicate if CIO reports
directly to CEO or president of the firm.
TOCFO Categorical variable to indicate if CIO reports
to a finance executive (to determine bias
towards
Financial Considerations)
____________________________________________________________________________________
ISOUT represents IS Outsourcing Policy (OP) - higher ISOUT implies lesser
reliance on internal IS department
and greater reliance on external sources: both inside as well as outside
the organization. IS Productivity
is measured by the "Information Productivity Index" (or IPI) as presented
in the table of Computerworld
Premier 100. IPI, which has been described as a better measure of IS
productivity than more traditional
measures, is expected to represent the return on management of IS
(Strassmann 1990,1994).
1 A. Structural Form of the Model
The structural model is based upon the equations (1) and (2) defined before.
OP = A0 + A1*F + A2*B (3a)
ISP = B0 + B1*F + B2*OP (3b)
where,
OP = Outsourcing Policy (Endogenous
Explanatory)
ISP = IS Performance (Endogenous)
F = Financial
Considerations (Exogenous)
B = Business Strategy (Exogenous)
Ai = Parameters that are assumed to have specific values
Bi = Parameters that are assumed to have specific values.
2. Determination of Indices to Be Used in the Analysis
Factor analyses was used to determine the
underlying dimensions of exogenous variables F and B.
Table 2 exhibits the summary of data and indices obtained from the factor
analysis done for determining
the underlying indices.
Table 2. Data and Indices Obtained by Factor Analysis
___________________________________________________________________________________
Factor Analyses
--------------------------------------------------------------
Data Reference Index
____________________________________________________________________________________
Financial Considerations(F)
Return on Technology A Performance index
NIIT
SAIT
Technology cost structure
ITGP
ITNP
LDSE
EAPS A Long-term index
Business cost structure A Business cost index
CSSA
CSTA
REOA
TLSE A Short-term index
Business Strategy(B)
Strategic Position of IS B Position index
TOCEO
TOCFO
Relative Importance of IS B Competitive index
ISBDGT
BDGTCHG
ISPEMP
____________________________________________________________________________________
Factor analysis (Reference A) of the ten indices for
Financial Considerations listed in Table 1
reveals
that four factors together explain 68.1 percent of the total variance.
While the first and third factors
represent the subcategories of return on technology and business cost
structure, the second factor represents the
longterm effect of IT investments on financial performance, and the
fourth factor explains the link between
increased total liabilities and corresponding decrease in return on
assets [due to investments in IT], and
hence represents the short-term effect of IT investments on financial
performance.
Factor analysis (Reference B) of the five Business Strategy indicators
supports their categorization into
the two classes that were based on intuition.
3. Two Stage Least Squares Method for Determining the Parameters of
the Structural
Equations
Two stage least squares method was used to determine the
parameters of the structural equations. Ordinary
stepwise regression was done on the equations of the reduced model.
Parameters from this
regression were then used to derive the estimated values (EISOUT) of the
endogenous explanatory variable
(ISOUT index of the OP variable). The estimated values were used for the
next stepwise regression to
determine the parameters of the structural equation (3b) for the other
endogenous (non-explanatory)
variable.
Tables 3 and 4 depict the results of the two regression
analyses. In both cases, the F-value suggests
that the regression is significant at .01 significance level. The
R-square values indicates that in both regressions
over twelve percent variance is explained by the explanatory variables.
Table 3: Determinants of IS Outsourcing (ISOUT)
____________________________________________________________________________________
Explained Variable
_______________________________________
Explanatory Variables a,b,c % IS Budget Spent Outside IS Department
____________________________________________________________________________________
R-Square .1421
F valuec [2,72]
5.96
Sig F .004
Constant 138.93
A. Financial Considerations
Performance 91.23
(1)
Business cost -77.50
(2)
____________________________________________________________________________________
a Only the explanatory variables selected by stepwise method up to the
0.05 level of significance are included in this table
b For each variable, data are values of the regression coefficients;
numbers in parentheses indicate step when entered into the
corresponding regression.
c Values in brackets are corresponding degrees of freedom for F
coefficient of the equation.
Table 4: Determinants of IS Productivity (IPI)
____________________________________________________________________________________
Explained Variable
_______________________________________
Explanatory Variables a b Information Productivity Index
____________________________________________________________________________________
R-Square .1210
F valuec [2,81]
5.574
Sig F .005
Constant .2775
A. Financial Considerations
Business cost -.068086
(1)
Short-term -.056296
(2)
____________________________________________________________________________________
a b c Same as for Table 3
Results in Tables 3 & 4 suggest the relative importance of
Financial Considerations in determining
the IS Outsourcing Policy and the firm's IS Productivity. Of the
Financial Considerations, performance
seems to
positively influence the IS outsourcing decisions, while higher business
cost structure has a negative
effect. Business Strategy considerations do not seem to influence IS
Outsourcing Policy. Further, IS
Outsourcing Policy does not seem to have any direct influence on IS
Productivity. However, these results
could have been influenced by the limitations inherent in this kind of study.
4. Conclusion
Limitations: Considering the fact that this is the first study of
its kind which is based on data that is
available for only one year, several limitations must be observed.
Firstly, interactions between the various
variables were ignored. Secondly, since IPI and other related measures
have only been used recently for
Computerworld rankings, a longitudinal study was not feasible. Thirdly,
within the scope of this study, the data
for IS budget spent outside IS department was used as a surrogate for the
IS Outsourcing Policy. Any delays
between the allocation of IS budget and the resulting effect on IS
performance were not considered for analysis.
Finally, it is difficult to ascertain the generalizability of this study
to companies that were not in the Premier
100 list.
The results of the study are interesting in several
respects. First, based on the results, it can be argued
that the new measure of IS performance used by Computerworld, although
intended to give due emphasis to
management of IS, hasn't adequately addressed its objective. Second,
inconsequential significance of the
Business Strategy indices could be attributed the indirect influence of
IS investment on the organizational IS
performance measures such as IPI. One approach for overcoming this
limitation could involve using mediating
variables that could represent process-level or function-level
contribution of the IS function. A second
approach could involve comparison of the market value of the IS function
with alternative [internal or external]
options available to the organization. Further research in IS
outsourcing can reveal more reliable and valid
findings by taking into consideration the findings of this study and the
limitations mentioned herein.
[an error occurred while processing this directive]