رشد سرمایه انسانی و تخریب: تاثیر زاد و ولد در منسوخ شدن مهارت

This paper investigates the effect of exogenous fertility shocks on physical- and human capital accumulation in an Overlapping Generation (OLG) framework. Negative shocks in fertility cause a relative scarcity in the future labor market and stimulate human capital accumulation. In turn, this allows the introduction of new technologies at a faster rate, thereby accelerating the obsolescence of the human capital held by older workers who become less employable. The model can thus explain why, after a period of declining fertility, labor market participation of older workers slumped.

Are older employees less productive? During the last decades, the labor participation of older employees has declined severely (for instance, see Hurd, 1997), although life expectancy has increased. A lot of older employees are unemployed or have been offered an early retirement arrangement by the firm.1 What could be the reason for the exodus of older workers and why would firms cooperate?2 This paper illustrates that fertility shifts can affect the employability of older workers. A mechanism is examined, with an endogenous rate of depreciation of human capital, which may explain why, in periods of a decline in fertility, skills of older employees get obsolete. As a result of a decline in fertility, labor will become scarce relative to physical capital, which boosts the return on human capital and stimulates human capital accumulation. For the youngest generations, human capital will be increased by receiving more education. Working generations accumulate human capital in particular by learning-by-doing and they have the incentive to supply more labor, thereby increasing their knowledge of existing technologies. A higher level of education of the young adults induces adoption of new technologies. When new generations are better trained, the average human capital level of the labor force increases, which eases introduction of more sophisticated technologies. Moreover, the higher wage level induces introduction of new technologies that are more capital-intensive. Upgrading of technologies used implies that knowledge of older versions of the technology become partially obsolete. For instance, the availability of sophisticated software packages for empirical analysis makes extensive knowledge of programming languages like Fortran unnecessary. Because of the partial obsolescence of knowledge of older technologies, human capital is subject to depreciation: the human capital effectively used, will decline. Older workers have much knowledge of (the version of) the technology that gets obsolete, and their productivity falls behind after the introduction of new versions of technology. The human capital of the youngest generations consists primarily of the technologies that are just introduced. As a result, the labor market position of older employees deteriorates.3 The role of the adoption of new versions of technologies in the decline of the labor participation of older employees is still under debate. There is indirect evidence, however, that the rate of technical change is affected by demographic shifts. Cutler et al. (1990) investigated the relation between productivity growth and labor force growth and found a strong negative relation during the period 1960–1985. That the level of technical change is higher when fertility declines, is the result of the assumption that the quality and not the quantity of the labor force is the decisive factor in the process of technical change. This is in contrast to Van Imhoff (1988) who suggested that a high population rate means a high influx of recently educated people and a higher rate of technical change. It is in contrast as well to the notion that an increased market size makes innovations more profitable, which would imply a positive relation between population and the number of ideas (and thereby technological change). As noted by Jones (1995), this argument is at odds with the postwar growth figures. Though the increased market size led to a higher level of R&D, this did not result in higher rates of economic growth. Young (1998) solved this puzzle, allowing for rents of innovators dissipated by either increased quality improvement or increased entry. An expansion of the labor force gives a level effect of the economy: the size of sectors as well as the variety of products increase. In the literature, two types of depreciation of human capital are distinguished: technical and economic skill obsolescence. In case of technical obsolescence, skills get lost because of lack or insufficient use of skills or because of the natural aging process. Economic obsolescence means that the value of the human capital is affected, for instance due to technological and organizational developments. See De Grip and van Loo (2002) for an overview of the literature. In this paper, the human capital of older workers is a resultant of formal schooling during childhood, and the net effect of learning-by-doing and skill obsolescence during the working period. Although in empirical studies, the relative importance of human capital from formal schooling and learning-by-doing is analyzed (for instance, Rosenzweig, 1994), in economic theory, both methods of human capital accumulation are not modelled together. Azariadis and Drazen (1990) have different productivity levels of younger and older workers. They use a two-period Overlapping Generation (OLG) model in which agents supply labor in both periods and invest in human capital in their first period, to increase their labor efficiency in their second period of adulthood. They do not model human capital accumulation by learning-by-doing nor a depreciation rate of human capital. Kremer and Thomson (1998) treat differences of young and older workers, but concentrate on the effects of imperfect substitutability of human capital of young and old workers in production. Galor and Tsiddon (1997) investigate the interaction between human capital and technological progress and focus on the distribution of human capital. Endogenous reasons for depreciation of human capital are hardly modelled. An exception is Sala-i-Martin (1996), who considers a depreciation of human capital for old generations dependent on the human capital level of the young generation. In the study of Ahituv and Zeira (2000), older workers might be confronted with a decline in their income-capacity because of technical change. In their model, old and young workers compete with each other. Young workers learn to use new technologies, whereas old workers stick to—the less-efficient—old technologies. The present paper adds a new aspect to the existing literature by stressing on the relationship between demographic shifts and skill obsolescence. It investigates the effects of fertility shifts on the accumulation (i.e., skills obsolescence) of physical capital and human capital of young and old workers. This is done with a modification of the two-period OLG model of Azariadis and Drazen (1990). In the model, endogenous growth originates from the transfer of knowledge and skills accumulated by preceding generations. Parents care about the educational level of their offspring. Human capital of the young adults in the first period depends on the time invested by the parents and the human capital level of the parents in the preceding period. In the first period, a household can spend time for investment in education of their offspring or alternatively increase its own human capital by working with existing technologies. In the second period, the household consumes its savings and the wage earned in the second period of adulthood. The human capital level in the second period is an increasing function of the first period human capital and the proportion of time spent working in the first period, and is subject to depreciation. The size of the depreciation depends on the level of investment in human capital per child. The model enables an analysis of the changes in the supply of labor efficiency of young and old generations on the labor market under demographic shifts. A negative shock in fertility leads to higher investment in human capital per child by formal schooling, but also, as long as depreciation of human capital is modest, to higher total investment in education of the descendants. The human capital level of new generations increases, which induces a higher rate of adoption of technology. This will make knowledge of older (versions of) technologies partially obsolete and the depreciation of knowledge of older workers accelerates. As a result, the labor position of older employees deteriorates relative to the situation without the demographic shift. Furthermore, the social return to schooling is lower than the private return when fertility declines. Households do not take into account the effect of schooling on the human capital of older workers. This paper is structured as follows. Firstly, the model is presented in Section 2. In Section 3, I analyze the dynamics of the model. In Section 4, the effects of a fertility shock on human capital accumulation are investigated.

In this paper, the dynamics of human capital with fertility shocks are investigated. An Overlapping Generations model is used where younger workers have accumulated human capital by formal schooling and working generations by learning-by-doing. The human capital of working generations is subject to depreciation. The rate of depreciation is the result of the arrival of new technologies. The arrival rate of new technologies depends on the change in the average human capital level of the employees, as a higher level of human capital induces adoption of better technologies. As older employees do not find investment in learning new technologies profitable by assumption, the arrival rate is determined by the level of investment in human capital per child. It is found that a positive shock in fertility results in lower investment of human capital per child. Hence, a decline in fertility leads to a higher education level of the next generation, which induces adoption of new technologies, thereby making skills of the older generation obsolete. This mechanism makes the arrival of new technologies (and thereby the level of skill obsolescence) endogenous. This relation between fertility and depreciation of human capital adds a new aspect to the literature on skill obsolescence. This relation also helps to explain why, after the period of declining fertility, a decline in the participation rate of older workers has occurred. In sectors where the adoption rate in technology is high, human capital might decline in absolute terms as well. If wages are sticky, older workers are too expensive compared to younger workers in these sectors. Because of the external effect of schooling on the human capital of the older employees, the level of investment in education is suboptimal. It might be optimal to subsidize schooling of older employees and to propagate permanent education, to improve their position on the labor market. Future research should investigate the optimal public policy. Furthermore, the effects of heterogeneity in ability should be investigated: upgrading of technologies affect higher- and lower-educated workers differently. Because higher-educated workers adopt new technologies easier, and the human capital of workers is subject to depreciation, this might help to explain the worsened position on the labor market of lower-educated workers.