implications of the Lewis turning point for the labor supply, and hence, economic growth. Investigating the trends of rural-to-urban migration and more general labor market changes shows that agriculture no longer serves as a pool of surplus labor; rather, rural workers’ migration to, and settlement in, urban areas has become irreversible and inevitable. Section III discusses the outdated policy implications of the Todaro paradox in the circumstances of the Lewis turning point. This suggests that there is no reason to expect a come-and-go pattern of rural labor migration and that an appropriate policy choice would be to push forward with urbanization by transforming migrant workers into urban residents. Section IV shows the urgency and feasibility of including migrant workers in the urban social security system from the viewpoint of government policy orientation. Section V concludes with some policy suggestions.
II. The irreversibility of rural?Curban migration
There has been much disagreement about whether (or when) China has reached (or will reach) its Lewis turning point (Cai 2008a, 2008b). This is not all that surprising considering
that, according to Lewis (1972) and other authors (for example, Ranis and Fei 1961), there are, in fact, two turning points as a consequence of development in a dual economy. The period when the growth of labor demand outstrips that of labor supply―and hence the wage rate of unskilled workers begins to rise―is the first Lewis turning point. At this point, the agricultural wage is not yet determined by the marginal productivity of labor, so a productivity differential between the agricultural and modern sectors still exists. Subsequently, the point at which wages in the agricultural and modern sectors are determined by their respective marginal productivities of labor and are equal to each other is the second Lewis turning point, which is also referred to as the commercial point. When an economy reaches the latter point, it is no longer a dual economy. This section intends to discuss the first of these two turning points.
The Lewis turning point is not, and should not be, a black-and-white watershed that distinguishes between two stages of development; rather, it is a transitional period bridging them―or it can be viewed as the starting point of a new historical trend in the course of economic development (Minami 1968). In this regard, while 2004 was a significant year
which marked the turning point, the longer period around it is the focus of this study. In what follows, we examine the different characteristics of the labor market before and after the turning point.
During the stage long before the Lewis turning point, due to the fact that there was a large pool of surplus labor in agriculture and the marginal productivity of labor in that sector was very low, the shift of workers from agricultural to non-agricultural sectors did not impact agricultural
production―that is, labor migration at this stage did not cause a significant change in the production mode of agriculture. On the other hand, because non-agricultural sectors during this stage absorbed the transferred labor force only marginally and sporadically, and since urban authorities frequently dispelled migrant workers when urban labor markets came under pressure (Cai et al. 2001), the agricultural sector still served as a pool of surplus labor. With the arrival of the Lewis turning point, however, circumstances have shown some fundamental changes, which we explain as follows:
First, the methods of agricultural production have changed in response to the massive and unremitting outflow of workers. Such outflow, given its large scale and steady growth, has
accelerated agricultural mechanization and modernization and has pushed a transformation of agricultural technological change from being of the labor-use type to the labor-saving type. Examining the changes in agricultural mechanization clearly shows this trend. During the first three decades of economic reform, the total power of agricultural machinery has been strengthened, and even with the enlarged base, the growth has shown no sign of easing in recent years. What is even more notable is the changed composition of different-sized agricultural tractors and tractor-towing machinery. In the period 1978?C98, when there was a surplus of labor in agriculture, the average annual growth in the capacity of large and medium-sized tractors was 2 per cent, while that of small-sized tractors was 11.3 per cent. In the period 1998?C2008, as the mass labor force shifted from agricultural to non-agricultural sectors, stronger demand emerged for labor-saving technological advances; the capacity of large and medium-sized tractors increased by 12.2 per cent annually and that of small-sized tractors decreased by 5.2 per cent. Changes in the growth rates of different sizes of tractor-towing machinery showed a similar trend, with the average annual growth rate of large and medium-sized
tractor-towing machinery increasing from zero per cent in the period 1978?C98 to 13.7 per cent in 1998?C2008, whereas the average annual growth rate of small tractor-towing machinery declined from 12.1 per cent to 6.9 per cent in the same period. As a result of falling labor inputs and rising physical inputs in production, China’s agricultural capital?Clabor ratio―which is denoted by the ratio of physical inputs to labor inputs―has risen rapidly since 2004 (see Figure 1). According to the theory of induced technological changes (Hayami and Ruttan 1980), this labor-saving tendency during the rapid process of agricultural mechanization is the natural result of the ultimate abatement of the surplus labor force in agriculture. It is hardly surprising that the total factor productivity (TFP) of the agricultural sector has also witnessed a rapid rise during the same period―increasing by 38 per cent between 1995 and 2008, with a sharp rise after 2004 (Zhao 2010).
Second, urban demand for migrant workers has become increasingly rigid. As the result of a demographic transition, during which the process of the urban age structure changes faster than the rural age structure, economic growth in urban sectors depends heavily on labor supply through migration. And yet, as shown in Table 1, while the total number of migrant
Urbanization with Chinese Characteristics
