An Empirical Exploration Of The Inverted U Problem

Since the reform and opening, between the urban and rural income the gap has been widening. Urban and rural income ratio has risen more than tripled, also the GINI coefficient exceed the international warning line. Factors which affect the income gap have several reasons, the most important factor is economic growth and financial development. Most scholars believe that the relationship between income inequality and economic growth is the so-called inverted U hypothesis, but there are a considerable number of people who believe inverted U curve between income inequality and economic growth is not established, it is reason that people like to study.Wagner M(2015), Wagner M, Hong SH(2015) pointed out that powers of an integrated process are, clearly, not integrated processes of any order,Moreover, they also verified this conclusion. but due to insufficient, we can not personally verify this conclusion. However, the author will do re-integration of Chinese data,to study and exploration income distribution.By studying statistics theory and empirical methods of domestic and foreign economists, income inequality will be retesting.This paper will establish a nonlinear and co-integration regression model with squared terms, collecting per capita GDP as indicators of economic growth time series from1978 to 2014 years, and panel data of 31 provinces from 2002 to 2014 years. For the part of time series, this article use three different estimation methods co-integration analysis test,including FM- OLS, D-OLS and ordinary least squares. First of all, for per capita gross domestic product, expenditure, total exports, total investment, and the GINI coefficient(AJ & JC) will get on descriptive analysis and unit root test, because the data are non-stationary, the co-integration regression model will be established, and need to Cointegration and coefficient test. About panel data of 31 provinces, the author will describe the data statistical analysis of indicators, and use FM-OLS, D-OLS and GMM of these three methods. In accordance with time series and panel data. The descriptive statistical analysis and mode result can be seen that they exist the nonlinear inverted U curve relationship between income distribution and economic growth(financial development).