Measuring Efficiency of Manufacturing Industries in Pakistan: An Application of Dea Double Bootstrap Technique

• Author: HAFIZ GHULAM MUJADDAD AND HAFIZ KHALIL AHMAD

Abstract. This is the very first study, which analyzes technical efficiency and its sources for the large scale manufacturing industries (LSMI) of Pakistan through DEA double bootstrap technique. First, we applied bootstrapped DEA technique for measuring bias-corrected technical efficiency scores by utilizing four inputs and one output. Finally, we employed the bootstrapped truncated regression model for determining the sources of technical efficiency. It is found that industries should reduce their size as there is evidence of diseconomies of scale in our results. Average wage as a measure of workers skill level has positive impact on technical efficiency. Finally market size does not have any significant impact in regression analysis.

Keywords: Technical efficiency, Large scale manufacturing industries (LSMI), DEA double bootstrap, Truncated regression, Pakistan

1. INTRODUCTION

   In today’s world the globe consists of the borderless economy and each and every entity should be prepared to accept the challenges of this change if they want to play a major role in businesses and remain competitive. An entity must be efficient if it wants to stay in businesses. So the performance measurement is necessary for this purpose as the efficiency is the only criteria for organizations to remain in business. It is essential for firms, organizations or industries to reach at their optimal level in order to compete with their business competitors all over the world. It is the requirement of every country to see that its organizational performance is good with high efficiency and productivity in order to achieve its targets. Traditional measures of performance are sales, worker turnover, share prices and exports etc. However, these measures do not reflect the complete picture of a firm’s performance.

   In contrast to these measures, efficiency is a more comprehensive measure as it is based on both the inputs and outputs.

   Our study is the first one that analyzes efficiency of large scale manufacturing industries using two-stage bootstrapping technique. Basically, performance measurement is to compare the efficiency of different units, compare the present level of efficiency to previous level of efficiency, comparison of actual efficiency scores to planned efficiency levels, comparison of different geographical zones or performance can be measured by comparing the efficiency of entities functioning under the similar conditions (Wholey and Hatry, 1992). The common criteria for measuring technical efficiency (TE) are to maximize output or profit and minimize the cost. Under certain circumstances, TE is determined as the skill of an industry to produce. An organization or industry is considered as technically efficient if it is producing maximum output from a specific amount of inputs or it is producing a given amount of output by using the minimum amount of inputs. The aim of the producers is to minimize the wastages.

   Measuring efficiency is one of the main aspects of today’s world. Every entity is curious about its performance and is afraid of failure which may not only harm the industry’s reputation but may also shake the investors’ confidence in management. Performance evaluation has two basic features: i) it shows the effect of past decisions, and ii) it shows the formation of financial structure of any firm or industry. It is well known that basic purpose of efficiency evaluation is to determine whether industries are employing their resources in the most efficient way (Duzakin and Duzakin 2007). Performance is not a simple concept but it is relatively close to productivity and efficiency. The concept of efficiency indicates that a firm or industry can produce by utilizing the minimum resources or inputs like capital, labor and other expected inputs and is able to remain competitive over the long period of time (Mayes et al., 1994).

   Different researchers have defined efficiency in different ways as Koopmans (1951) defines TE as “There is a possible point in the commodity space which is known as efficient point whenever an increase in one of its coordinates (the net output of one kind) can be obtained only at the expense of decline in some other coordinates (the net output of other kind)”. Another definition of efficiency involves equating the inputs and outputs of an organization with that of its peers which are performing well. These peers are estimated in relation to the specific objective where it can be measured based on the output maximization, profit maximization or cost minimization (Thanassoulis, 2001).

   The concept of measuring efficiency of producing units was initiated by Farrell (1957). There are two basic techniques which can be used for measuring efficiency; parametric and non-parametric. Aigner et al. (1977), and Meeusen and Broeck (1977) developed the parametric approach (stochastic frontier analysis, SFA) first. Linear programming models, which are also known as non-parametric approach, of Charnes et al. (1978) and Fare et al. (1985) which are based on convexity assumption are famous with the name of data envelopment analysis (DEA). Both approaches have some limitations: SFA necessarily requires specification of functional form and assumptions regarding distribution of the error term. In contrast, DEA does not require these conditions.

   It is assumed in DEA that decision making units (DMUs), which are producers (economic agents), have control over the discretionary variables but Ouellette and Vierstraete (2004) and many others have justified that non-discretionary inputs are also present in every sector and therefore, these environmental variables must be used in DEA model.

   Simar and Wilson (2007) show that earlier studies that involved models such as DEA based on two-stages of production processes were defective due to their failure in describing the data generating process (DGP). Therefore, these approaches are invalid due to the presence of serial correlation in the estimated efficiency scores. The authors identified many problems of these approaches which combine DEA with Tobit regression models. Therefore, they introduced DEA double-bootstrap approach. This approach enables construction of confidence intervals for estimated efficiency scores and let identify its determinants.

   The present study is aimed to evaluate the technical efficiency (TE) of large scale manufacturing industries (LSMI) of Pakistan. LSMI are chosen because it is the major part of the industrial sector which has much importance in Pakistan. Industrial sector is the third largest sector after services and agriculture sectors. Manufacturing sector contributes 13.5% of Gross Domestic Product (GDP) and absorbs 14.1 percent of total employed labor force. The productivity and performance of LSMI sector has much importance for sustained growth and development of the country as LSMI comprises more than fifty percent of the industrial sector.

   However, it is not sufficient to measure the technical efficiency only without determining its sources. This study measures efficiency scores of LSMI and also assesses its determinants as Pakistan is ranked 133 among 148 in global competitiveness index (the Global Competitiveness Report 2013-14; Schwab (2014)). There is no study in Pakistan that has estimated bias-corrected efficiency scores of manufacturing sector and assessed its sources. This will be the first study to evaluate the technical efficiency and its determinants by applying the DEA double bootstrap. In this study industries are also regrouped for making them comparable.

   The remaining of the study is ordered as follows: Review of literature is given in section 2. Section 3 provides methodological framework and describes sources of data. Empirical results of manufacturing industries are interpreted in Section 4. Section 5 consists of conclusions and policy recommendations.

2. REVIEW OF LITERATURE

   Many studies have been done for measuring the performance of industries. But almost every study adopted the common technique of DEA (especially in case of Pakistan) for measuring the efficiency of different sectors including manufacturing sector. First, we review the literature related to efficiency, then related to Pakistan.

   LITERATURE RELATED TO EFFICIENCY

   Mahadevan (2002) analyzed the performance of productivity growth of 28 manufacturing industries of Malaysia over the period of 1981 to 1996. He applied the data envelopment analysis (DEA) to compute Malmquist index of total factor productivity (TFP) growth and technical change, change in technical efficiency and change in scale efficiency were decomposed from Malmquist index. They used three variables (capital, labor as inputs and value added as output). They found that the non-ferrous metal industry obtained the highest TFP growth i.e. 3.7 percent and petroleum refineries obtained the lowest TFP growth i.e. -0.3 percent. They also found that the average weighted TFP growth was 0.8 percent; technical change was 0.3 percent; technical efficiency change was 0.5 percent; pure technical efficiency change was 0.4 percent and scale efficiency change was 0.1 percent.

   They argued that low TFP growth was due to the small gains in both technical change and technical efficiency, with industries operating close to optimum scale.

   Baten et al. (2006) analyzed the technical efficiency of selected manufacturing industries of Bangladesh by applying the stochastic frontier production approach over the period from 1981/1982 to 1999/2000. They covered the selected 3-digit census factories. They included three variables (value added, capital and labor) in their research. They applied two alternative distributions to model the: the truncated normal distribution and the half normal distribution. They found that estimated technical efficiency for selected industries was 40.22% of potential output under the truncated normal distribution while it was 55.57% of potential output for the half normal distribution...