Eyeballing the clustering effect in China

By Michelle Lee Twan Gee

SMU Office of Research & Tech Transfer – In a well-known economic phenomenon called agglomeration economies, companies in the same industry that cluster together enjoy benefits that lead to higher productivity.

However, most studies into agglomeration economies focused on examining the influence of agglomeration economies on private companies in developed economies such as the United States. Hence, that phenomenon has largely been observed and analysed only in developed economies.

Would the effects of agglomeration economies be the same for developing economies and for state-owned enterprises in these economies, wondered Professor Li.

To investigate this, the Assistant Professor of Economics, together with her SMU colleague Professor Li Liyao and Professor Liu Shimeng at Jinan University’s Institute for Economic and Social Research, carried out a large-scale study for a complete list of twenty-nine manufacturing industries in China.

Their findings were set out in a recent research paper titled ‘Attenuation of Agglomeration Economies: Evidence from the Universe of Chinese Manufacturing Industries.’

The unique features of Chinese manufacturing

Why China, in particular?

Explained Professor Li: “China is quite different from other economies, especially in terms of its manufacturing sector. The Chinese manufacturing sector is very big and fast growing. Also, in contrast to developed economies, the educational level of its labour force tends to be low and it is still catching up in terms of the innovation capacity.

 “With the level of skill and innovation within the Chinese manufacturing sector, interaction between companies can be quite different. In addition, a unique feature of China is the large presence of state-owned enterprises.”

China’s unique manufacturing sector makes it interesting to examine whether agglomeration economies also hold true for China. In turn, the forces that influence agglomeration effects on Chinese enterprise will help to shed light on the economic forces facilitating China's dramatic economic growth, said Professor Li.

She added, “As the largest developing country in the world, evidence in the China context also help to verify whether the standard patterns of agglomeration economies also hold for the developing world, adding to existing literature on agglomeration economies in developing countries.”

How businesses benefit from having close neighbours

According to the theory of agglomeration economies, companies benefit from being located in the same neighborhood as other firms in the same industry. Specifically they benefit in three ways: through knowledge spillover, through labour pooling and through input sharing.

In knowledge spillover, firms that are located close to each other will talk to and learn from each other even if they are competitors. This knowledge spillover will ultimately lead to an increase in productivity for the firms in the area. An example of this is Silicon Valley.

The hallmark of this effect is that it attenuates very rapidly with distance. Knowledge spillover often requires face-to-face contact. This means that firms have to be located within close proximity of each other to enjoy the effect. Hence, if knowledge spillover is behind the agglomeration economies enjoyed by a particular industry, the decay or attenuation speed will be very rapid over distance. Firms that are physically far away from their nearest compatriots will not enjoy much knowledge spillover.

In labour pooling, when similar firms concentrate in one location, they will demand the same types of labour. By doing so they draw workers possessing certain sets of skills to the area. This, in essence, enhances the skill matching between the employers and the employees in the area, which raises labour productivity for firms in the industry in the locale. 

The third vector that gives rise to agglomeration economies is input sharing. When similar firms are sited in the same area, they will demand similar inputs. By locating close to each other, they will be able to reduce the transportation cost of shipping in the same types of inputs, thereby creating economies of scale for their raw materials. By this means, each firm’s productivity is enhanced.

Does it pay to be neighbours in China?

The research turned up some very interesting findings.

The first is that the benefits of clustering fall away very rapidly over distance in China, similar to what has been documented in developed economies.

As the paper highlighted, “We find that the agglomeration economies attenuate with geographic distance. The initial attenuation is rapid, with the effect of own-industry employment in the first kilometre significantly larger than the effect of employment further away.”

The study also found that the state-owned sector in China shows a weaker agglomeration effect than the private sector. This suggests that the agglomeration economies that state-owned enterprises enjoy owes more to input sharing – which reduces less over distance – than knowledge spillover or labour pooling.

The study also shed light on industry-specific patterns. Described the paper, “The estimated attenuation speed varies dramatically across industry types.”

The rate of reduction of agglomeration effects depends on which underlying forces are in action. For instance, industries that rely heavily on knowledge spillovers as the main agglomeration force would require close range face-to-face contact, which implies a rapid decay of agglomeration effects over distance. In contrast, industries that cluster mainly because of input-output linkages would see agglomeration economies decaying slowly over distance, even as the benefits of agglomeration extend over a larger geographical distance.

For example, the food production industry has the highest reliance on transportation and a high manufactured input share among all the industries. For this industry, the agglomeration effect is significant even when the firms are 10 to 20 km away from the geographic centre of activity. 

Meanwhile the electronic and telecommunications industry is the most reliant on knowledge sharing, and for this industry, agglomeration attenuates quickly, indicating that the forces of knowledge spillovers might operate at a smaller scope than input sharing.

The paper is important in several respects.

Said Professor Li, “We did this study to show evidence that the underlying regularities of agglomeration economies also hold true for China. And the data shows that it does hold true.”

The study also probed, for the first time in this research domain, how agglomeration economies differ for specific industry and the factors behind the attenuation pattern for each industry.

It also shed light – again for the first time in this research field – whether ownership type (that is, whether a firm is privately owned or state-owned) influences the degree of agglomeration economies. The paper suggests that for China, the answer is yes.

Back to Research@SMU Feb 2020 Issue