The reform of RMB exchange rate formation mechanism has been advanced constantly and RMB exchange rate flexibility also has been enhanced orderly since July 2015, when China adopted a managed, floating exchange rate regime with reference to a basket of currencies instead of fixed exchange rate regime being pegged to the US dollar. In August 2015, China once again carried out the reform aiming at improving the central parity rate mechanism of RMB exchange rate against US dollar. Since then, the two-way volatility of RMB exchange rate has been normalized. As the price of home currency in the international market, the exchange rate can produce price and wealth effects along with its fluctuations, thereby resulting in distributional effect. On the one hand, owing to differentiated trade goods produced and consumed between rural and urban residents, the RMB exchange rate changes play a role in trade structure, industrial structure, employment structure and economic growth, and further in urban-rural income distribution pattern, by affecting import and export trade. On the other hand, the RMB exchange rate changes can affect types & amount of currencies that are possessed by urban-rural residents, and then have an important effect on cross-border capital investment. Especially with the internationalization of the RMB and the two-way opening-up of the financial market, the marketization of RMB exchange rate will be further heightened, thus it is no doubt that the effect of RMB exchange rate changes on income distribution will be strengthened.

As an important aspect of unequal income distribution, urban-rural income gap has always been a major concern of all circles in China. Under the dual urban and rural economic background, this paper theoretically depicts micro mechanism concerning the effect of RMB exchange rate changes on urban-rural income gap. Then it applies panel threshold regression model proposed by Hansen(1999) with provincial data during the period of 1994 to 2016 and endogenous grouping to empirically investigate asymmetric and regional heterogeneous features of the effect of RMB exchange rate changes on urban-rural income gap in 28 Chinese provinces. Our results show that the effect of real effective RMB exchange rate changes on urban-rural income gap is not fixed, but depends on changes in regional per capita income and the degree of trade openness, thereby being featured by asymmetry and regional heterogeneity. Specifically speaking, the appreciation (depreciation) of real effective RMB exchange rate can reduce (expand) the urban-rural income gap in the provinces with low income level & low trade openness as well as middle income level & high trade openness, but will expand (reduce)the urban-rural income gap in the provinces with high income level and high trade openness. However, the changes in the real effective RMB exchange rate do not have an obvious effect on the urban-rural income gap in the provinces with middle and high income levels and low trade openness.

Based on the conclusions of this paper, we can get the following enlightenment. Firstly, when formulating policies narrowing down the urban-rural income gap, Chinese governments should look upon the distribution effect of exchange rate changes and take different measures in different regions, to respond to the shock of exchange rate changes to urban-rural income gap. Secondly, to narrow down relative labor productivity gap between urban and rural areas is one of the important directions to alleviate income gap. Thirdly, in the process of expanding financial opening-up and trade openness, local governments should pay attention to the improvement of rural financial development and trade environment, and change the situation that rural areas cannot or will not enjoy the dividends of financial opening-up and trade openness to a greater extent, thereby providing conditions for narrowing down urban-rural income gap.

The contributions of this paper are as follows:firstly, the existing literature investigates urban-rural income gap mainly from the perspectives of macroeconomic development, policy differences and historical legacy, while this paper focuses on exchange rate changes, which would enrich the previous research; secondly, we utilize the panel threshold approach and further explore the regional heterogeneous feature of the effect of RMB exchange rate changes on urban-rural income gap through endogenous grouping according to the threshold value, which is helpful to examine the income distribution effect resulting from real exchange rate changes more comprehensively & objectively; thirdly, the theoretical and empirical analysis of this paper helps us to understand the intrinsic mechanism and external manifestation of the effect of RMB exchange rate changes on urban-rural income gap, and then provides some enlightenment for governments to make relevant policies to reduce urban-rural income gap.