Developing countries need sector-level infrastructure investment to replicate China’s rapid growth

16 September 2021

Adobe Stock 263302388
The accepted knowledge that strong individual companies and greater competition drives growth in developing economies is being challenged by a new study from an international team of researchers led by Oxford University and Jilin University in China.

They analysed growth patterns of firms during China’s 1998 to 2014 development push. The results suggest that sector-level infrastructure investments were essential in driving growth in employment and productivity, and that company-level growth followed from that. Firm-level investments would not have the same impact because individual firms could not prosper without a thriving sector.

Sectors are more than mere collections of firms in the same line of business, they are interconnected systems sharing sector-specific infrastructure, technology, and a labour force, according to the researchers at the University of Oxford (UK), Chemnitz University of Technology (Germany), University of Waterloo (Canada), Jilin University (China) and University of Duisburg-Essen (Germany). The study found that sector-level development is a significant predictor for individual firm success indicating that to drive economic growth investment should go towards training, technology, and to communication and transport networks, not to individual firms.

“The fortunes of an individual company may rise and fall, but the labour force and infrastructure create the environment for effective competition and growth,” said Professor Torsten Heinrich, associate of the Oxford Martin Programme on Technological and Economic Change and lead author on the study. "The correlation of productivity and output growth clearly only arise at the sector level. The firm-level dynamics don’t work the same way. However, sector-level growth still predicts firm-level growth, so the two are connected.”

Sectors are more than collections of firms, they are interconnected systems sharing infrastructure, technology, and a labour force

Sector-level correlation laws between employment growth, output growth, productivity and many other variables have been identified since the 1940s. High-growth sectors (in terms of output and income) tend to also lead in productivity and employment growth. The current study finds that many of these correlation laws are not present at the level of individual firms within sectors.

"This shows that it is not enough to support specific firms or stimulate growth in particular fields", says co-author Professor Shuanping Dai from the Centre for China Public Sector Economy Research at Jilin University. "While more research on other developing countries is needed, it is fair to say at this point, that development policy needs to prioritise supporting infrastructure to help sector ecosystems to develop. Successful individual firms will rise from that broader based investment."

The researchers also find that firm-level variables follow distributions with heavy tails, that is, distributions that are dominated by very large observations; firms with extremely high profits (or losses), productivity, or size. Professor Jangho Yang, co-author and associate of the Oxford Martin Programme on Technological and Economic Change, adds: "Statistical methods need to take this into account. OLS regressions, for instance, will produce misleading results; a more sophisticated robust regression with, e.g., Cauchy distributed errors must be used."