From Efficiency Analyses to Policy Implications: a Multilevel Hierarchical Linear Model Approach

Abstract

This paper examines the key factors that influenced the cost efficiency of 7,633 Vietnamese manufacturing firms during 2010–2016 via a hierarchical linear modelling (HLM) approach. The main reason for using HLM in this case is that observations in the same group may not be independent from each other (e.g. firms operate within the same city), and some variables may not vary across those observations. Although most of the findings are consistent with previous studies, the statistical power of our HLM model is higher than that of the traditional single-level analysis, suggesting that HLM can provide better analytical insights. The results further indicate a case for cities or provinces pursuing different policies aimed at improving the performance of their local firms.

Keywords:

• Efficiency
• hierarchical linear modelling (HLM)
• Vietnam
• manufacturing

JEL CLASSIFICATION:

• D24
• L60

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 Details of the SFA estimations regarding the cost efficiency of the sampled 7,633 Vietnamese manufacturing firms are discussed in Ngo et al. (2019a).

2 We are thankful for the anonymous reviewer for pointing this out. Since our research objective is to propose a better approach using HLM for the second-stage regression of efficiency and performance analysis that can deal with hierarchical/nested data, the straightforward comparison in Table 3 is justified. We leave further HLM analyses for future studies.