Estimation & Decomposition of Productivity Change in Food, Beverages & Tobacco Products Using Frontier Approaches.

AuthorRoy, Prasanta Kumar

Introduction

Together with the tobacco industry, the Food and beverages sector is a large source of manufacturing output and employment. However, productivity growth of this sector remains very low as it faces a conflux of challenges such as climate change, changes in food supply and demand and imbalances in the governance of food production systems, food price volatility, and food security (Roy & Das, 2018). The objective of this paper is to illustrate how we can use frontier estimation methods to obtain estimates of total factor productivity growth (TFPG), and decompose it into various sources such as technological change (TP), technical efficiency change (TEC) and economic scale change (SC) of the 4-digit manufacturing industries of food, beverages and tobacco products in India during the period from 1998-99 to 2017-18, and during two sub-periods of 1998-99 to 2007-08 and 2008-09 to 2017-18- the pre-and post-financial melt-down periods (the years are also said to be pre-and post-financial breakdown periods respectively). The 4-digit manufacturing industries of food, beverages and tobacco products considered in our study are: production, processing and preserving of meat and meat products [1010(1511)], processing and preserving of fish and fish products [1020(1512)], processing and preserving of fruit and vegetables [1030(1513)], vegetable and animal oils and fats [1040(1514)], dairy products[1050(1520)], grain mill products [1061(1531)], starches and starch products [1062(1532)], bakery products [1071(1541)], sugar [1072(1542)], cocoa, chocolate and sugar confectionery [1073(1543)], macaroni, noodles, couscous and similar farinaceous products [1074(1544)], other food products n.e.c. [1079(1549)], prepared animal feeds [1080(1533)], distilling, rectifying and blending of spirits; ethyl alcohol production from fermented materials [1101(1551)], wines [1102(1552)], malt liquors and malt [1103(1553)], soft drinks; production of mineral waters [1104 (1554)] and tobacco products [1200 (1600)].

Literature Review

Fare, Grosskopf, Norris and Zhang (1994) took the Malmquist index of TFP change, defined in Caves, Christensen and Diewert (1982), and described how we can decompose the Malmquist TFP change measures into various components, including technological change (TP) and technical efficiency change (TEC). They also showed how these measures could be calculated using distances measured relative to DEA frontiers. Ray (1997) used a non-parametric method of DEA to measure Malmquist productivity index for manufacturing sector in the different states in India for the period 1969-84. The measured Malmquist productivity index is decomposed to separate the contribution of technological change (TP), change in technical efficiency (TEC) and change in scale efficiency (SC). The analysis depicted that in most of the states productivity decline is due to technological regress. Ray (2002) examined the impact of reforms on efficiency and productivity in the manufacturing sector of Indian states for the period 1986-87 through 1995-96. Using data envelopment analysis, the study noted an improvement in TFPG in most of the states during the reforms period (1991-92 to 1995-96). The study showed that improvement in technical efficiency as well as faster technological progress had contributed to the observed acceleration in productivity growth. Mahadevan (2001, 2002) used both stochastic frontier approach and DEA separately to calculate the TFPG of Malaysian manufacturing industries during 1981-1996. He used the same data set to make comparison between the two approaches and concluded that both methods demonstrated a decline of TFPG after 1990, increasing contribution of technological progress (TP) and declining contribution of technical efficiency change (TEC). Kumar (2004) decomposed total factor productivity growth (TFPG) in industrial manufacturing in 15 major Indian states for the period 1982-83 to 2000-01 using non-parametric linear programming methods. He decomposed TFPG into efficiency and technological changes and level of biasness in technological change. The resulting information is used to examine whether the post-reform period shows any improvement in productivity and efficiency in comparison to the pre-reform one. Joshi and Singh (2010) measured the TFPG and identifies its sources through applying a non-parametric DEA-based MPI (Malmquist productivity index) approach. The productivity growth was decomposed into technical efficiency change (TEC) and technological change (TP). Heshmati and Kumbhakar (2010) in their study used panel data on Chinese provinces and identified a number of key technology shifters and their effect on technological change (TP) and TFPG. Using input-output data from the Annual Survey of Industries for the period 1970-71 through 2007-08, Deb and Ray (2014) compares the pre-and post-reform performances of Indian manufacturing in terms of total factor productivity growth (TFPG). They used Data Envelopment Analysis to construct a Biennial Malmquist Index of total factor productivity for individual states. Results showed that at the all-India level, total factor productivity growth rate in manufacturing is higher during the post-reform period. Although the majority of states experienced accelerated productivity growth, some states did experience decline in productivity after the reforms. Both before and after the reforms technological progress (TP) was the most important component of productivity growth in Indian manufacturing.

On the other hand, stochastic frontier model was pioneered by Aigner, Lovell and Schmidt (1977) and Meeusen and van den Broeck (1977). Nishimizu and Page (1982) was the first to propose de-composition of TFP into efficiency changes and technological changes using stochastic frontier approach. He used a deterministic translog production frontier to decompose productivity change in Yugoslavia into two aforementioned components. Bauer (1990) estimated a translog cost frontier using data on the US airline industry to decompose TFP growth into efficiency, technical progress, and scale components. Kaliranjan et al (1996) used stochastic frontier approach to decompose the sources of total factor productivity (TFP) growth into technological progress and changes in technical efficiency within the framework of the varying coefficients frontier production function using the Chinese provincial-level agricultural data covering the period 1970-87. By applying a flexible stochastic translog production function, Kumbhakar and Lovell (2000), Kim and Han (2001) and Sharma et al (2007) decompose TFP growth into four components: technological progress, changes in technical efficiency, changes in allocative efficiency and scale changes. Kim and Han (2001) applied a stochastic frontier production model to Korean manufacturing industries to decompose the sources of total productivity (TFP) growth into technological progress (TP), changes in technical efficiency (TEC), and changes in allocative efficiency (AEC), and scale changes (SC). Empirical results based on data of Korean manufacturing industries from 1980 to 1994 showed that productivity growth was mainly driven by technological progress, that changes in technical efficiency had a significant positive effect, and that allocative efficiency had a negative effect. They suggested that specific guidelines are required to promote productivity in each industry, and provided additional insights into understanding the recent debate about TFP growth in Korean manufacturing. Using the stochastic frontier production function and the decomposition method of total factor productivity made by Kunbhakar and Lovell (2000), Liu and Huang (2009) studied the total factor productivity of China's steel industry during 1981-2003. They found: capital and TFP play more important roles for output growth while the contribution of labor input is negative. Of all the factors for the growth of TFP, scale economy contributes the most, followed by allocation efficiency of resources. Though technical efficiency has been keeping pace with TFP, it contributes little to TFP growth and thus the improvement of technical efficiency is the key element for improving the efficiency of China's steel industry. Roy, Das and Pal (2017) decomposed the sources of TFPG of the thirteen 2-digit manufacturing industries as well as total manufacturing industry in West Bengal into technological progress (TP), changes in technical efficiency (TEC), economic scale effect (SC) and allocation efficiency effect (AEC) during the period from 1981-82 to 2010-11, during the pre-reform period (1981-82 to 1990- 91) and post-reform period (199192 to 2010-11) by using a stochastic frontier approach. According to their estimated results TFPG of almost all the 2-digit manufacturing industries in West Bengal have declined during the post-reform period which is mainly accounted for by the decline in technological progress (TP) of the same during this period. The change in allocation efficiency component, however, found that resource allocation in almost all the 2-digit industries in West Bengal has improved during the post-reform period. Njuki, Bravo-Ureta and O'Donnell (2018) estimated a stochastic production frontier model using U.S. state-level agricultural data incorporating growing season temperature and precipitation, and intra-annual standard deviations of temperature and precipitation for the period 1960-2004. They used the estimated parameters of the model to compute a TFP index that has good axiomatic properties. They then decomposed TFP growth in each U.S. state into weather effects, technological progress, technical efficiency, and scale-mix efficiency changes. Roy & Das (2018) applied stochastic frontier approach to decompose the sources of total factor productivity growth (TFPG) of the manufacturing industries of food, beverages and tobacco products in India into technological progress...

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT