Firm productivity/efficiency and its determinants – evidence from MSMEs in Vietnam and two other Southeast Asian countries

This thesis aims to measure firm-level labour productivity and technical efficiency using different techniques such as Data Envelopment Analysis (DEA) and Stochastic Frontier Analysis (SFA). It then analyses the factors that drive firm technical efficiency or productivity, especially the inter-relationship between firm efficiency/productivity, corruption, government assistance, innovation, and gender. The study focuses on Micro, Small and Medium-sized Enterprises (MSMEs), which are considered a lifeblood of economies. This thesis consists of three studies that make use of two different datasets of MSMEs. The first two studies use biennial survey data of manufacturing MSMEs in Vietnam covering four waves between 2009 and 2015. This data resulted from collaboration between Danish and Vietnamese agencies. The third study utilises World Bank cross-country data for manufacturing MSMEs in Vietnam, Indonesia, and the Philippines in 2009 and 2015.  

The first study investigates the inter-relationship between corruption, government assistance and firm efficiency for MSMEs in Vietnam. The study treats corruption, government assistance, and firm efficiency as endogenous variables in order to control for feedback effects between them. There exist only a small number of firm-level studies on corruption and firm efficiency. In addition, this study investigates the role played by government assistance in this relationship. Two contending hypotheses are tested in the analysis. The first suggests that corruption enhances firm efficiency by helping to “grease-the-wheels”. Under this hypothesis corruption aids efficiency by overcoming bureaucratic obstacles. The second hypothesis, referred to as “sand-the-wheels”, suggests that corruption reduces efficiency. The study finds a strong nexus between firm efficiency, bribery, and government assistance. The effect of bribing on firm efficiency is significantly positive, supporting the “grease-the-wheels” hypothesis. The study also finds that government assistance is effective in improving firm efficiency. The relationship between   bribery and government assistance is found to be bidirectional, implying that they reinforce each other. The study is careful to place corruption in its broader economic context. As such, despite its positive effect on firm efficiency in Vietnam, this study does not conclude that corruption is to be encouraged.  

The purpose of the second study is to empirically analyse the links between gender, innovation and firm labour productivity in Vietnamese MSMEs in the manufacturing sector over the years 2009–2015. Many studies confirm that innovation and R&D lead to increased productivity at firm and national levels. The evidence on the role played by gender in this nexus is mixed. A variant of the endogenous switching regression model and the Blinder-Oaxaca decomposition method is employed in the analysis. While other such studies tend to use cross sectional data, a distinctive feature of the study is the use of panel data. The study finds that innovation increases firm labour productivity. While the research does find a gender gap in innovativeness in favour of male owners/managers, this gap is small. Moreover, female-controlled firms are not significantly less productive than male-controlled firms. The results of this study contribute to a small but important body of literature on MSMEs analysing the link between gender, innovation, and firm labour productivity in emerging economies. Given that the number of females starting or running new businesses is higher than males in developing countries, with the biggest difference seen in Vietnam, it follows that female entrepreneurship is playing an important role in enhancing economic growth in emerging economies. The analysis provides some useful suggestions for formulating entrepreneurship-related policies.  

In the last study, the differences in technical efficiency of MSMEs in three Southeast Asian countries are highlighted. The study then explores the determinants of firm-level technical efficiency. The novelty of this study is that technical efficiency scores of firms are measured against a metafrontier, and therefore they are comparable between different countries that use distinct production functions. A two-step stochastic frontier approach is employed for the purpose of the analysis. The results show that there are no significant gaps in firm efficiency between Vietnam, Indonesia, and the Philippines. As for the sources of firm technical efficiency, various factors appear to affect efficiency scores. Firms become less efficient when they age. Firms only improve their efficiency level when they grow their size to more than 35 employees. Female-owned firms underperform in comparison with their male counterparts. Foreign-owned firms do not outperform firms without foreign ownership. Exporters are less efficient than firms only serving the domestic market. Hence, the data do not support the learning-by-exporting hypothesis. Firms receiving bank loans and firms providing formal training programs are not as efficient as firms without bank loans and firms without training, respectively.