In transition economies such as Vietnam, the diverse organizational forms of enterprises can be classified as state-owned, private, and foreign-owned enterprises. This column uses regression analyses to examine the impact of agglomeration economies or their sources on the productivity of different types of firms in Vietnam. The findings reveal that the productivity of state-owned enterprises is unrelated to agglomeration economies, whereas that of private and foreign-owned firms is likely to increase in at least one type of agglomeration economy through various channels.

Introduction

Vietnam has transitioned from a centrally planned economy to a market economy since 1986 when it adopted the Doi Moi economic reform policy. Consequently, a substantial influx of foreign direct investment stimulates an export-oriented manufacturing sector. The 1999 Law on Enterprises institutionalized ownership rights and the freedom to conduct business (Vu-Thanh 2017). Meanwhile, the 2007 accession to the WTO led to the unification of the Law on Enterprises for private firms and state-owned enterprises (SOEs), and the Law on Investment for foreign-owned firms and SOEs (Vu-Thanh 2017). Therefore, in Vietnam, SOEs, foreign-owned firms, and private firms coexist under the same rule. However, SOEs are linked to the government, foreign-owned firms are associated with foreign nations, and private firms may have evolved from traditional village-based industries. In this context, two intriguing questions arise: Whether agglomeration economies differently affect the productivity of these three firm types and which source of agglomeration economies is related with the productivity of those firm types.

Gokan, Kuroiwa, and Nakajima (2019) examined whether differences in firm characteristics influence the relationship between firm productivity and the magnitude of agglomeration economies, which may include localization and urbanization economy. According to Gokan, Kuroiwa, and Nakajima (2019), localization economy is associated with a cluster of firms within a firm’s own industry, whereas the urbanization economy is associated with some clusters of industries. Beaudry and Schiffauerova (2009) stated that “a diversified local production structure gives rise to urbanization (diversification) externality or Jacob’s externalities.” Jacobs (1969) argued that industry diversity is important for innovation. In short, our first question can be rephrased as whether agglomeration economies are related to an industry’s cluster or the number of clustered industries.

Moreover, Gokan, Kuroiwa, and Nakajima (2019) examined whether differences in firm characteristics influence the relationship between firm productivity and the sources of agglomeration economies as the second question. Marshall (1920) suggested three sources of agglomeration economies: sharing inputs, labor pooling, and knowledge transfer (Rosenthal and Strange 2004). Thus, we decompose the effects of agglomeration economies into three sources: interindustry transactions, labor pooling, and knowledge spillovers.

Data and Methodology

We applied the cluster detection procedure developed by Mori and Smith (2014) to identify significant cluster areas and to distinguish cluster and noncluster areas for each industry. This method is closely related to the cluster-identification procedure used in epidemiology to detect disease clusters.

We generated an index of localization economies by identifying whether a firm is located within a cluster in its own industry at the district level. We then develop a new index of urbanization economies by calculating the number of clustered industries in each district, using Mori and Smith’s (2014) method. This index measures the degree of diversity of clustered industries. Beaudry and Schiffauerova (2009) showed that the mainly employed indices of urbanization economies are the Hirschman–Herfindahl index and Gini index. These indices do not identify clusters during the index calculation process; therefore, their meaning is unclear. Nevertheless, our index was generated using only the clustered areas of each industry. Thus, our index reveals not only the diversity of clustered industries, but also the magnitude of firm concentration.

To identify the areas with clusters, we use data on the number of workers, location (district level), ownership type, and industrial classification at the four-digit level in the manufacturing sector under Vietnam Standard Industrial Classification (VSIC) in the fourth Establishment Census of 2012 conducted by the General Statistics Office (GSO) of Vietnam.

The productivity used in this study is the total factor productivity obtained by estimating a Cobb–Douglas production function using data on location (district level), ownership type (i.e., state-owned, private, or foreign-owned), establishment code, industrial classification at the two-digit level in the manufacturing sector, number of workers, capital, and value-added of the enterprises in the fourth Establishment Census by the GSO of Vietnam.

On the sources of agglomeration economies, we determined interindustry transaction relationships utilizing the 2012 input–output table provided by the GSO of Vietnam, labor pooling utilizing the 2012 Vietnamese household living standard survey, and interindustry knowledge spillovers using the results of a questionnaire survey included in the Establishment Census.

The number of state-owned, private, and foreign-owned firms in the manufacturing sector are 612, 45,069, and 5,008, respectively. Most private firms are micro and small enterprises, which contribute to their large number.

To control for unobserved heterogeneity across districts and industries, we included prefectural and industrial fixed effects in the regression analysis. The two-digit VSIC code was used to classify industries.

Total Factor Productivity of Three Types of Firms

The mean total factor productivity of state-owned, private, and foreign-owned firms are 0.246, 0.232, and 0.289, respectively, with only foreign-owned firms significantly outperforming private firms. This contradicts Ramstetter and Ngoc (2013)’s findings, which showed that SOEs have significantly higher productivity than private firms in Vietnam.

Total Factor Productivity in Urban and Nonurban Area

Defining urban area as an area with an above-median number of clustered industries, we compare the mean total factor productivity across firm types and two-digit industries. We find no difference in the mean total factor productivity of SOEs between urban and nonurban areas, whereas the mean total factor productivity of private and foreign-owned firms is significantly higher in urban areas. Thus, the urbanization economy primarily benefits private and foreign-owned firms. Furthermore, the majority of industries with significantly higher mean total factor productivity in urban areas than in nonurban areas are light industries such as food, tobacco, and textiles. These findings contradict those of Henderson (2003), who found that urban economies are effective for the high-tech industry.

Localization Economies and Urbanization Economies

We estimate the relationship between firm productivity and localization economies using regression analysis. Using all available data, we demonstrate that localization economies increase the productivity of firms in Vietnam. However, we find that SOEs do not benefit from localization, whereas private- and foreign-owned firms do.

We estimate the relationship between firm productivity and urbanization economies using regression analysis. Using all the available data, we demonstrate that urbanization economies do not increase the productivity of firms in Vietnam. We find that neither SOEs nor private firms benefit from urbanization economies, whereas urbanization economies increase the productivity of foreign-owned firms.

Interindustry Transaction, Labor Pooling, and Knowledge Spillovers

We estimate the relationship between firm productivity and the three sources of agglomeration economies using regression analysis. Using all available observations, we found that firms with transaction relationship have higher productivity in clusters. Moreover, SOEs do not benefit from any source of agglomeration economies. Meanwhile, agglomeration economies are effective for the productivity of private firms primarily through interindustry transactions and partially through labor pooling. The agglomeration economies for foreign-owned firms are mainly derived from knowledge spillovers.

Conclusion

First, the relationships between agglomeration economies and the productivity of the three types of firms are distinct. In Vietnam, the productivity of state-owned firms is unrelated to agglomeration economies, whereas that of private and foreign-owned firms is likely to related with localization economies. The productivity of only foreign-owned firms is related with urbanization economies.

Finally, the relationship between the source of agglomeration economies and the productivity of the three types of firms are also distinct. For private firms, agglomeration economies are effective through inter-firm transactions, whereas those for foreign firms are effective through knowledge spillover. In other words, there is no correlation between the productivity of state-owned or private firms and the diversity of the clustered industries. This result corresponds to Vietnam's weak agglomeration economies, as noted by Ketels et al. (2010). The weak urbanization economies are consistent with Henderson's (2003) finding that only high-tech industries experience urbanization effects.

Author's Note:

This column is based on: Gokan, Toshitaka, Ikuo Kuroiwa, and Kentaro Nakajima. 2019. “Agglomerations Economies in Vietnam: A Firm-Level Analysis.” Journal of Asian Economics 62: 52–64. https://doi.org/10.1016/j.asieco.2019.03.002

References

Beaudry, Catherine, and Andrea Schiffauerova. 2009. “Who's Right, Marshall or Jacobs? The Localization versus Urbanization Debate.” Research Policy 38 (2): 318–337.

Duranton, Gilles, and Diego Puga. 2004. “Micro-Foundations of Urban Agglomeration Economies.” In Handbook of Regional and Urban Economics, Volume 4, edited by J. Vernon Henderson and Jacques-François Thisse, 2063-117. Amsterdam: Elsevier.

Gokan, Toshitaka, Ikuo Kuroiwa, and Kentaro Nakajima. 2019. “Agglomerations Economies in Vietnam: A Firm-Level Analysis.” Journal of Asian Economics 62: 52–64. https://doi.org/10.1016/j.asieco.2019.03.002

Henderson J., Vernon. 2003. “Marshall’s Scale Economies.” Journal of Urban Economics 53: 1-28.

Jacobs, J. 1969. The Economy of Cities. New York: Vintage.

Ketels, Christian, Nguyen Dihn Cung, Nguyen Thi Tue Anh, and Do Hong Hanh. 2010. Vietnam Competitiveness Report 2010. Hanoi/Singapore: Central Institute for Economic Management.

Marshall, Alfred. 1920. Principles of Economics. 8th Edition. London: Macmillan.

Mori, Tomoya, and Tony E. Smith. 2014. “A Probabilistic Modeling Approach to The Detection of Industrial Agglomerations.” Journal of Economic Geography 14(3): 547–88.

Ramstetter, Eric D., and Phan Minh Ngoc. 2013. “Productivity, Ownership, and Producer Concentration in Transition: Further Evidence from Vietnam.” Journal of Asian Economics 52: 56-76.

Rosenthal, Stuart S., and William C. Strange. 2004. “Evidence on the Nature and Sources of Agglomeration Economies.” In Handbook of Regional and Urban Economics, Volume 4, edited by J. Vernon Henderson and Jacques-François Thisse, 2119–71. Amsterdam: Elsevier.

Vu-Thanh, Tu-Anh. 2017. “The Political Economy of Industrial Development in Viet Nam.” In The Practice of Industrial Policy, edited by John Page and Finn Tarp. Oxford: Oxford University Press.

Acknowledgement

I would like to thank Ikuo Kuroiwa and Kentaro Nakajima for useful comments.

* Thumbnail photo: Vietnam, Ho Chi Minh (Saigon) seen from a plane, activity centre (Daniele Schneider / Photononstop / Getty Images)
** The views expressed in the columns are those of the author(s) and do not represent the views of IDE or the institutions with which the authors are affiliated.