The importance of small and medium-sized enterprises (SMEs) to economies in Asia is well known. They account for over 95% of all businesses, a third to half of aggregate output, and the majority of enterprise employment (Vandenberg, Chantapacdepong, and Yoshino 2016).
We also know that SMEs do not have an easy life. They struggle to get established, face a higher failure rate than large firms, and lack access to key inputs such as finance. Finding ways to increase their survival rate and growth is important for expanding private sector activity in Asia’s developing economies. Sustaining enterprises requires that they are competitive; competitiveness, in turn, is based on productivity.
What then contributes to SME productivity? There are many factors: the technology they use, the way the production process is organized, the skill of the owner and/or manager, and, of course, the quality of the workforce. High worker productivity contributes to high enterprise productivity.
Linking human capital and productivity
Our recent research tested the basic intuition linking the human capital of the enterprise workforce with enterprise productivity (Vandenberg and Trinh 2016). We wanted to see whether we could establish this basic link and determine whether a skilled and educated workforce is as important for SMEs as it is for large firms. We also sought to understand differences between countries. Our results offered up the expected results but also threw up some interesting differences between countries.
We used data on more than 4,000 enterprises from the World Bank’s Enterprise Surveys database. The firms were from five countries: the People’s Republic of China (PRC), Indonesia, Malaysia, Thailand, and Viet Nam. Over 1,000 firms each were included from the PRC and Thailand, and between 500 and 750 firms for each of the other three countries. Small firms were defined as employing up to 100 workers, whereas medium-sized firms employed 100–249 workers and large firms 250 workers or more. We understand that firm-size categories vary between countries but we imposed common definitions to be able to pool the data.
We defined human capital in two ways. One was the average education level of the workforce and here we made a distinction between enterprises with an average of less than 10 years and those with 10 years or more. Ten years is normally sufficient to complete junior high school or high school. We also defined human capital in terms of whether the enterprise provided formal training for its workers. Firms that trained their workers were making a specific attempt to raise the quality of the workforce. More detailed information on training (quality, type, duration) would have been useful but was not available in the dataset.
We then constructed variables for labor productivity and capital per worker, and included variables for enterprise size, age, industry, in-country location, and country. We used econometric techniques to test the significance of the relationships between the key variables and labor productivity, our dependent variable.
The details of the techniques are provided in our paper (Vandenberg and Trinh 2016). For the reader who is familiar with econometrics, we can indicate here briefly that we ran the basic regression models and then also interacted country and enterprise size variables with the human capital variables in further models. We ran additional estimates using instrumental variables to address concerns about endogeneity and to allow us to suggest a relationship of causality—not just correlation. We checked for robustness.
Confirming our intuition
The results confirmed our intuition. We found that firms with a more educated workforce and those that offered formal training had higher labor productivity. Furthermore, both of our human capital inputs had an independent impact on productivity. This suggests that enterprises need not choose between hiring educated workers and training existing workers; both can be done, simultaneously, as part of a broad strategy to raise productivity.
Training and education were also significantly correlated with each enterprise size category. That is, human capital benefits small firms as it does medium-sized and large firms. However, human capital inputs had a greater impact on the productivity of larger firms than it did on that of smaller ones.
We used a large, cross-country sample to ensure that we were generating valid results. However, we also broke down the analysis by country to see whether one or two countries were driving the overall results. We found that the human capital variables did not impact productivity in the PRC and are not sure why. Enterprise training is high in the PRC (higher than in the other countries), and it may be that, because most enterprises train, differences in training (very little) could not be related to differences in productivity (which might be much greater). Thailand showed the strongest impact of human capital. In Viet Nam, education impacted productivity but training did not, whereas the reverse was true in Indonesia.
Aside from human capital, we also found significant productivity differences based on enterprise size. Not surprisingly, small enterprises were the least productive and large enterprises were the most productive. As expected, we also found that higher capital per worker was associated with higher productivity.
Not only finance
Access to finance is often the focus of attention when discussing the barriers to SME development. Policy recommendations also tend to focus on designing or enhancing credit support programs. However, our research shows that education and training have a significant bearing on enterprise productivity. Therefore, programs to raise human capital should also be part of SME support strategies in Asia.
_____
References:
Vandenberg, P., and L.Q. Trinh. 2016. Small Firms, Human Capital and Productivity in Asia. ADBI Working Paper No. 582. July. Tokyo: Asian Development Bank Institute.
Vandenberg, P., P. Chantapacdepong, and N. Yoshino, eds. 2016. SMEs in Developing Asia: New Approaches to Overcoming Market Failures. Tokyo: Asian Development Bank Institute.
Comments are closed.