MEASUREMENT OF TECHNICAL EFFICIENCY IN HIGHER TECHNICAL EDUCATION IN INDIA

Abstract

Higher education is of vital importance to educational systems in contemporary societies. Moreover, knowledge-based industries are now the major drivers of economic growth and development in a fast changing globalised world. Furthermore, globalization and integration of economies has led to greater economic opportunities for countries with good levels of education (Carnoy (1999), Tilak (2001)). There has been a significant growth in the number of higher educational institutions in India over the past six decades. In particular, there has been a phenomenal expansion of higher technical education in the country that is regarded as without parallel anywhere else in the world (Tilak (1999)). The most significant achievement has been the development of 7 world-class Indian Institutes of Technology (IITs), Indian Institute of Science, Bangalore (IISc) and 20 National Institutes ofTechnology (NITs). To evaluate the competency of institutions that provide higher technical education in India (in terms of number of students produced, quality of students' training, etc.), it is necessary to assess the economic efficiency of these institutions. More specifically, it is necessary to study their technical efficiency by taking the educational inputs and outputs into consideration. The term technical efficiency (TE) can be defined as "the ability and willingness offirms to produce the maximum possible quantity of output with a specified endowment ofinputs, given the technology and the environmental conditions that surround them" (Kalirajan and Shand, 2000). Precisely, a firm is said to be technically efficient when it realizes its technical potential for the given set ofinputs and technology. The measurement of technical efficiency of higher technical educational institutions is important in order to: (a) learn the scope for increasing educational outputs, given the existing inputs, (b) improve the accountability of these institutions and (c) make relevant policy decisions regarding continuance of public funding towards higher technical education. Technical efficiency of higher educational institutions has been analyzed extensively in developed countries such as Australia, Germany, Japan, South Africa, U.K. and U.S.A. by using the Stochastic Frontier Production Function methodology or Stochastic Frontier Analysis (SFA) andDataEnvelopment Analysis (DEA) (Abbott and Doucouliagos (2003), Izadi et al. (2002), Kaneko (1997), Robst (2001), Taylor and Geoff (2002), and Vink (1997)). However, no rigorous study has been carriedout to measure the technical efficiency of higher educational institutions in India. In particular, the technical efficiency of higher technical educational institutions in India has not been studied so far. This thesis examines the technical efficiency of important higher technical educational institutions in India, namely, 7 IITs, IISc and 17 (out of 20) NITs, by using the SFA technique forpanel data. More specifically, Battese and Coelli's SFA model as well as fixed effect model (FEM) and random effect model (REM) approaches towards SFA are employed to measure the technical efficiency of the above institutions. The dataset used in the study consists of 4-year 'unbalanced' and 'balanced' panel data pertaining to IITs and IISc, for academic years 2001-02 to 2004-05, and 3- year 'unbalanced' and 'balanced'panel data pertaining to NITs, for academic years 2002-03 to 2004-05. The data utilized in the study has been gathered from the annual reports of the above institutions, which are available at Department ofSecondary and n Higher Education, Ministry ofHuman Resource Development (MHRD), Government ofIndia, New Delhi, as well as the institutions' administrative offices. To calculate the technical efficiency of the above institutions using the SFA technique, the study utilizes four output variables and three input variables. The output variables used are: Total Research Publications (TRP), which measures the research output of an institution in a given year; Total Enrollment (ENR), which measures the teaching output of an institution in a given year; Degrees Awarded (DA) by an institution in a given year; and Effective Placement (EP) of an institution in a given year. The input variables used are: Academic Staff (AS), Non-Academic Staff (NAS) and Library Stock (LS). The parameter estimates of the Battese and Coelli's model of SFA have been obtained by using ordinary least squares (OLS) and maximum likelihood estimation (MLE) methods with the aid of the FRONTIER programme, which accommodates panel data as well as 'time-varying' technical efficiency. For the FEM and REM approaches towards SFA, the MATLAB package has been run using the panel data technical efficiency programme of Sickles (2003) that can accommodate several statistical and econometric models, such as Battese and Coelli's model, FEM, REM, and DEA, in a single programme. The study demonstrates that technical efficiencies ofeducational outputs vary widely across sample units (i.e. IITs, IISc and NITs). It is also shown that technical efficiencies pertaining to educational outputs of the above institutions can be 'timevariant' (increasing or decreasing TE over time) or 'time-invariant' (i.e. constant TE over time). The study postulates that the technical efficiency pertaining to an educational output of an institution can be low due to inefficient utilization of educational inputs/resources, which suggests that the concerned educational output of an institution can be raised through better application of existing inputs. in Furthermore, the study highlights the need for strengthening the know-how (concerning higher technical education) of the above institutions, so that these institutions can exploit the full potential of the existing educational inputs.