Redai dili (Dec 2023)
Intensity of Rubber Planted and Population Urbanization: A Spatial Econometric Analysis Based on Panel Data of Cities and Counties in Hainan Island
Abstract
Population urbanization is the core of new urbanization in China and is an important task in the long term. Based on mediation effect models, spatial panel data econometric models, and coupling coordination models, this study attempts to identify the direct and indirect effects of the proportion of rubber-planted areas and the indications of population urbanization using panel data of counties on Hainan Island, explaining the mechanism. The results show that rubber plantations dominated by state-owned farms weakly impacted farmers' income in 1992-2002 (first phrase), and the effects became positive in the period of smallholding expansions from 2003 to 2013 (second phrase) at a statistically significant level of 10%. However, the effects became negative between 2013 and 2020 (third phrase) due to the continuously low prices of natural rubber compared to the second phrase; the regions with higher share of rubber plantations had less local fiscal revenues per capita—especially in the third phrase—mainly because of the short industrial chain of rubber in producing regions. Shares of rubber-planted areas in the city or county (defined as rubber-planted intensity, RPI) were negatively related to proportions of staff in the non-private sector (SNS) and nonrural employees (NRE) compared to people with jobs and census registered population urbanization rates (PUR). The reversed connections between RPI and NRE became weaker over time; a 10-percentage point increase in PRI might have led to a 10.3-percentage point decrease in NRE in the first phrase, but it reduced to 3.3 percentage points in NRE with no statistical significance in the third phrase. Conversely, the negative correlation of RPI with SNS and PUR became stronger from the first to third phrase; the farmers' income and local fiscal revenue could ease the negative relationships between RPI and SNS or NRE, but they turned to enhance the relation of RPI with PUR in the second and third phrase. The evidence from Moran's I and Geary's c indices proved the existence of obvious spatial correlations in the distribution of natural rubber and farmers' income and indicated that GDP per capita and other variables are also spatially related. The estimated spatial panel data models with adjacent, geographical and economical distance matrices indicated that the direct influences of RPI on SNS, NRE, and PUR were almost negative in all phrases, but the spillover and total effects can be positive in some situations, such as in the second phrase with adjacent and geographical distance matrices of PUR; this means that an increase in rubber-planted areas may promote nonfarm job market of adjacent regions mainly due to rural labor mobility. The coupling coordination degrees between the intensity of rubber planted and population urbanization of Haikou and Sanya reach 0.96 and 0.87, respectively, in the second and third phrases—much higher than those of other regions. Compared to the first phrase, the margin of increases in the degrees of coupling coordination in the second and third phrases are reducing from the east to west in Hainan Island. The key routes to mitigate the negative links are to improve labor productivity and extend the industrial chain of natural rubber along with the nonfarm job creation, and Hainan's population urbanization should consider many other possible angles across the island.
Keywords
