Can regional integration promote industrial green transformation? Empirical evidence from Yangtze River Delta Urban Agglomeration

Abstract

This paper analyzes the spatial–temporal evolution of regional integration and industrial green transformation and empirically investigates the impact of regional integration on industrial green transformation using prefecture-level panel data of cities in the Yangtze River Delta Urban Agglomeration (YRDUA) from 2006 to 2018. The major findings are as follows: (1) during the observation period, the level of regional integration rose in general, with regions with a high integrated level expanding northward, southward, and westward in the YRDUA. Industrial green transformation has the property of path-dependent and a spatial lock-in pattern of “high in the east, low in the west.” (2) There exists a considerable U-shaped relationship between regional integration and industrial green transformation; most cities in the YRDUA have crossed the inflection point in terms of integrated development. Regional integration has a positive spillover effect on the industrial green transformation of the surrounding areas, which diminishes as geographic distance increases. When the level of regional integration becomes excessive, it also impedes surrounding areas’ industrial green transformation. (3) Regional integration can affect industrial green transformation by promoting economic scale expansion, industrial structure upgrading, and technological progress. At this moment, the economic scale is the most significant mediating factor. With the advancement of regional integration, technological progress will become critical. This paper provides a rationale for understanding industrial green transformation from the perspective of regional integration and enriches theoretical and empirical studies on ecological civilization construction and green development in urban agglomerations.

Appendices

Appendix 1

Yangtze River Delta Urban Agglomeration (41): Nan**g, Wuxi, Xuzhou, Changzhou, Suzhou, Nantong, Lianyungang, Huaian, Yancheng, Yangzhou, Zhenjiang, Taizhou, Suqian, Hangzhou, Ningbo, Wenzhou, Jiaxing, Huzhou, Shaoxing, **hua, Quzhou, Zhoushan, Taizhou, Lishui, Hefei, Wuhu, Bengfu, Huainan, Maanshan, Huaibei, Tongling, Anqing, Huangshan, Chuzhou, Fuyang, Suzhou, Luan, Bozhou, Chizhou, and Xuancheng.

Appendix 2

Calculation steps of regional integration by relative price variance method: (i) List K kinds of commodity’s relative price indices in city i and city j of year t ${P}_{it}^{k}$, ${P}_{jt}^{k}$ and then calculate the absolute value of their natural log with the equation: ${P}_{ijt}^{k}=\mathrm{ln\;}({P}_{it}^{k}/{P}_{jt}^{k})=\left|\mathrm{ln\;}{P}_{it}^{k}-\mathrm{ln\;}{P}_{jt}^{k}\right|$. (ii) Since the fixed effect brought by specific commodity category will lead to a certain degree of system bias, we need to use De-mean method to eliminate it, in order to obtain the residual difference only related to segmentation factor. The equation is $\Delta {P}_{ijt}^{k}={P}_{ijt}^{k}-\overline{{P }_{t}^{k}}$. $\overline{{P }_{t}^{k}}$ is the average value of all ${P}_{ijt}^{k}$. (iii) Repeat the above steps to calculate the remainder of relative price margin for other kinds of goods. (iv) Calculate the variance of $\Delta {P}_{ijt}^{k}$, which is $Var(\Delta {P}_{ijt}^{k})$. Variance in each administrative unit constitutes time series data, enables us to directly observe the evolution of variance along with time, and to test the changing trend of market integration by using self-moving law of time series. Eighty-eight thousand five hundred sixty relative price variances can be obtained based on the retail price index of 9 goods in 41 cities over 13 years; among them, we have 820 (${C}_{41}^{2}$) groups of paired cities. (v) Combine these relative price variances with the data of same province by year to obtain market segmentation index of each city (segin). Finally, the square root of segin’s reciprocal is used to calculate the core variable of this paper: regional integration. Carrying on a logarithm analysis, the relevant equation is $inter=\mathrm{ln}(\sqrt{1/segin})$.