Title: Determinants of Efficiency in Least Developed Countries: Further Evidence from Nepalese Manufacturing Firms
Abstract: Abstract Using a translog stochastic production frontier and maximum likelihood econometric methods, we estimate and model the determinants of firm level efficiency in the Nepalese context. Our results are broadly in line with theoretical expectations. We find that large firms are more efficient and that a higher capital intensity leads to inefficiency. There is no statistical evidence to suggest that foreign participation leads to efficiency improvements. Also, we do not observe any link between export intensity and efficiency improvement. We find that higher protection leads to inefficiency. Overall, our results suggest that an outward looking industrial strategy, which relies on less intervention and permits the development of large-scale industries, is conducive to efficiency improvement in least developed countries (LDCs) like Nepal. Notes For example, Sharma et al. [ 2000 Sharma, K, Jayasuriya, S and Oczkowski, E. 2000. Liberalisation and Productivity Growth: The Case of Manufacturing Industry in Nepal. Oxford Development Studies, Vol.20, No.3: pp.205–21 [Google Scholar] ] do not find any link between industry protection and productivity growth, while UNIDO [ 2002 UNIDO 2002 Industrial Development Perspective Plan: Vision 2020 Prepared for HM's Government of Nepal, Ministry of Industry, Commerce and Supplies by the United Nations Industrial Development Organisation [Google Scholar] ] and Sharma [ 2004 Sharma, K. 2004. The Impact of Policy Reforms on Labour Productivity, Price Cost Margin and Total Factor Productivity. South Asia Economic Journal, Vol.5, No.1: pp.55–68 [Google Scholar] ] establish a link between protection and productivity growth. Under the scheme, all imported raw materials are stored in a warehouse under the bond. Exporters are required to deposit duties, which are refunded after producing the evidence of exports. The bonded warehouse scheme was initially introduced in 1988 to encourage garment exports, later it was also extended to other export-oriented industries. There are few industry-level analyses of productivity growth in the Nepalese manufacturing which suggest that even after liberalisation total factor productivity (TFP) continued to decline. A recent study by the United Nations Industrial Development Organisation [ UNIDO, 2002 UNIDO 2002 Industrial Development Perspective Plan: Vision 2020 Prepared for HM's Government of Nepal, Ministry of Industry, Commerce and Supplies by the United Nations Industrial Development Organisation [Google Scholar] ] indicates that out of 44 industries at the four-digit level, 23 recorded an absolute fall in TFP growth. Most of these are labour-intensive industries, producing mainly for export markets. However, these results should be taken with a great degree of caution due to the highly aggregated nature of data. For further discussion on the impact of liberalisation on trade intensity see Sharma et al. [ 2001 Sharma, K, Oczkowski, E and Jayasuriya, S. 2001. Liberalisation, Export Incentives, and Trade Intensity: New Evidence from Nepalese Manufacturing Industries. Journal of Asian Economics, 12: pp.123–35[Crossref] , [Google Scholar] ]. These investments have come in electronic goods assembly activities, food and beverages, garments and carpets, and the chemical sub-sector. Based on the OECD classification, firms are grouped into resource intensive, labour intensive, specialised supplier, scale-intensive and science-based industries in Table 1. Resource based industries include: food, beverages and tobacco, wood products, petroleum refining, non-metallic mineral products and non-ferrous metal. Labour intensive industries are: textile, jute manufacturing, carpets, apparel and leather, metal products and other manufacturing. Specialised industries are: non-electric machinery, electric machinery, communications equipment and semiconductors, while scale-intensive industries are paper and printing, chemical excluding drugs, rubber and plastics, iron and steel, ship building, motor vehicles and other transport equipment. Science-based industries include aerospace, computers and office equipment, pharmaceutical and scientific instruments [ Sharma, 2001 Sharma, K. 2001. Liberalization, Growth and Structural Change: Evidence from Nepalese Manufacturing,. Applied Economics, 33: pp.1253–61[Taylor & Francis Online] , [Google Scholar] ]. For a very small number of firms, some specific inputs are not used. This poses estimation problems for production frontier estimation as the logs of inputs are employed. The frontier specification includes dummy variables to account for zero input usage for some variables, this overcomes estimation bias problems associated with the use of zero inputs [ Battese, 1997 Battese, GE. 1997. A Note on the Estimation of Cobb-Douglas Production Functions when some Explanatory Variables have Zero Values.. Journal of Agricultural Economics, 48: pp.250–52[Crossref] , [Google Scholar] ]. The estimated increasing marginal products for the three inputs deserves some comment. First, the results only hold for input values around the sample data as the estimated frontier does not define the entire shape of the frontier for all input values [ Coelli, Rao and Battese, 1998 Coelli T Rao DSP Battese GE 1998 An Introduction to Efficiency and Productivity Analysis Boston: Kluwer [Crossref] , [Google Scholar] ]. Second, the estimates may be the result of model mis-specification. We experimented extensively with different measures for the inputs, including the number of employees for labour, combining local and intermediate inputs, including fuel, water and electricity and measuring capital without a capacity utilisation adjustment. Moreover, various other determinants of inefficiency were considered, including R & D expenditure, ERP and public sector involvement. Despite this experimentation the presented specification proved to be most superior in terms of the specification tests, meaningful estimated input elasticities, individual firm efficiency estimates and estimates of the inefficiency determinants. For a very small number of firms, some specific inputs are not used. This poses estimation problems for production frontier estimation as the logs of inputs are employed. The frontier specification includes dummy variables to account for zero input usage for some variables, this overcomes estimation bias problems associated with the use of zero inputs, see Battese [ 1997 Battese, GE. 1997. A Note on the Estimation of Cobb-Douglas Production Functions when some Explanatory Variables have Zero Values.. Journal of Agricultural Economics, 48: pp.250–52[Crossref] , [Google Scholar] ]. Additional informationNotes on contributorsEdward OczkowskiEdward Oczkowski and Kishor Sharma, School of Commerce, PO Box 588, Wagga Wagga, Charles Sturt University, NSW, Australia, 2678. [email protected]; [email protected]. The authors are grateful to Philippe Scholtes of the UNIDO for permitting us to use the survey data collected for the preparation of Nepal's Industrial Development Perspective Plan:Vision 2020, Kathmandu, October 2001 (Project NC/NEP/00/009). Comments from two anonymous referees are gratefully acknowledged.
Publication Year: 2005
Publication Date: 2005-05-01
Language: en
Type: article
Indexed In: ['crossref']
Access and Citation
Cited By Count: 31
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot