Determinant of Labor Income with Policy and Institutional Factors - Innovative Management in Information and Production - page 109

Bai and Qian ( 2009a )and Jiang and Zhang ( 2008 ) are the problem about how
to model the nonlinear variables. The simple linear model can find the linear
significance, but not the nonlinear significance, the difference stems from the
simple linear modeling methods.
Question 3: From our results, the influence of secondary and tertiary industry is
nonlinear and linear under policy perspectives but both linear under complex
perspectives. Our finding supports most of study about the significance of indus-
try structure, which adopt the linear model specification. But Fang ( 2011 ) finds
the influence of individual industry is significant, but when they merge different
industry together, the influence change from significance to insignificance. We
find the difference arises owing to the mechanism of secondary industry is
nonlinear according to our finding.

4 Economic Regions

In the classical economic region analysis papers, China is divided into three parts
(East, Middle, West). From our results, if the difference between regions is big
enough and the connections are small enough, the bandwidth of region should be
close to 0. But the bandwidth of classical division is about 0.5. We set the GDP
per capita of 2009 as the new division rule: setting the GDP per capita as our new
division rule is beneficial that the bandwidths of region is almost 0 now.

4.1 The First Economic Region

The province whose GDP per capita is above 25,000 RMB is the so-called the first
economic region, which consists of Shanghai, Beijing, Tianjin, Jiangsu, Zhejiang,
Guangdong, Neimenggu, Shandong, Liaoning, Fujian, and Jilin.
In the first economic region, technological progress is the determinant. It has a
nonlinear way considering the institutional factors; it has a linear way considering
the policy factors; it has a linear way considering the complex factors combined
the policy and institutional factors. Trade is the determinant as well, the influence
of which is linear. Considering the complex factors, the influence is nonlinear.
Secondary industry is the determinant and has a linear way under the policy
perspective, but it is not the determinant under the complex modeling way. On
the contrary, tertiary industry is not the determinant under the policy perspective
but changes to the determinant under the complex modeling method and has a
linear way.