Trade Liberalisation, Productivity, and Protection: The Case of an Open Developing Economy - Dataset

Abstract

This study uses the National Census of Manufacturing Establishments (NCME) of Nepal which is conducted by the Central Bureau of Statistics (CBS). The CBS is the main government organisation in Nepal responsible for the collection, management, and dissemination of statistical information. These censuses are conducted quinquennially. This study utilises data from the census year 1981/82 to 2018/19 which gives eight time points for the panel dataset. Censuses 1981/82 and 1986/87 cover manufacturing establishments using automachines irrespective of the number of persons engaged, whereas censuses from 1991/92 to 2018/19 cover establishments engaging 10 or more persons. Each census contains an average of 4000 establishments. Following international practices, these establishments are categorised into 66 different industries according to the four-digit Nepal Standard Industrial Classification (NSIC). Therefore, each industry comprises of a group of establishments. As each census provides a combined value for different variables for establishments that belong to the same industry, analysis is conducted at the industry level since establishments are not uniquely identified in the census data. Industries with just one or two establishments in any census are omitted from the dataset as they do not reveal their data because of confidentiality issue. The censuses cover only industries that are government-registered and operate within the geographic boundary of Nepal. Given different industrial classification revisions that have occurred during the study period, conversion charts are used to confirm that industries are correctly categorised according to the latest NSIC for each census year. Some industries with one or two observations during the study period are merged with their closest allies based on the nature of their activities in the census year. This makes sure that they belong to the same 3-digit NSIC classification. This study also utilises the United Nations Conference on Trade and Development (UNCTAD) – Trade Analysis Information System (TRAINS) database to extract tariff data via the World Integrated Trade Solution (WITS) portal. Extracted tariffs for each of the 4-digit 66 industries are effectively applied tariffs (AHS). WITS defines these tariffs as the lowest available import tariffs. The most favoured nations (MFN) tariffs are also used as an alternative measure. The tariff rates for each industry are matched to their respective census years.