Poverty Traps:Risk, Human Capital Transition And Vulnerability

Poverty is one of the most important topics concerned in the process of development all the time. Breaking the low-level equilibrium and escaping from the poverty traps has always been a dream of humankind. Thousands of scholars paid great attention to this issue and made discussion and research related to it. Aiming to further explore the mechanism of poverty traps and the resulting changes associated with household welfare, this thesis focus on the research of human capital upgrade induced by technology and human capital transition under shock and risk, resulting with the impact of country’s choice on technology selection as well as human capital accumulation and households’choice on access to their children’s education as well as human capital transition. Based upon these two kinds of choice, we also try to analyze household’s characteristics and factors of welfare and vulnerability resulted from the mechanism of poverty traps discussed above. Finally, we attempt to explore the impact of the heterogeneity of individuals on the effectiveness of implementation of the economic policy.In the third chapter, we attempt to analyze the relationship among induced technology, human capital upgrade and development traps through a macroeconomic model. We consider that the technical level is demonstrated by technology users so that the human capital accumulation can be induced by technology upgrade. The model assumes that the level of human capital in developing countries is proceeding with the upgrade process of technical level, which means that the step by step human capital cumulation should be induced by step by step technology upgrade. Based upon this consideration, we define a class of non-smooth and periodic production function, where human capital is a concave function of the investment in education. Within each upgrade stage, the potential maximum level of human capital is the upper limit of technical level during that stage.Through the R&D investment and technology upgrade, sustained economic growth is driven by human capital accumulation. According two different types of human capital accumulation, we obtain two different types of economic growth, the sustained growth mode and trembling growth mode, which can explain the existence of low-level equilibrium trap in developing countries. Through numerical simulation we also consider about human capital accumulation induced by two extremes of technology upgrade, the mode of sustaining induced technology and the mode of trembling induced technology. Conclutions are that technology update by blindly learning and imitating in developing countries may not be able to help them to catch up with the leaders. Because the gap between current human capital in developing countries and potential upper limit of technology in developed countries is too huge and as a result, human capital upgrade threshold is too big to make it fall into the low-level equilibrium. In order to get rid of the traps, developing countries can take methods to improve the rate of human capital accumulation and to choose a more appropriate level of technology for human capital upgrade.In the fourth chapter, we attempt to analyze the relationship among the risk, intergenerational human capital transition and poverty traps in a microeconomic model. Statistical results indicate that human capital accumulation is of great importance for a family to improve their persistent poverty. But in fact, compared with high-income families, low-income families are reluctant to allow their children to accept education. Based on Galor and Zeira (1993), we construct a microeconomic model to study the role of risk played in education decisions and explain the problem of persistent poverty in low-income families. Based upon the analysis of this model we get three hypotheses regarding the relationship among the cost of human capital investment, expected return and risk of human capital investment, opportunity cost and risk of human capital investment. By using eight years’ survey data of the China Health and Nutrition Survey (CHNS), we empirically test the hypotheses through unbalanced panel discrete choice model and quantile regression model. The main conclusions are summarized as follows. Firstly, results from sub-sample processing and Chow-test show that the aspiration of human capital investment is positive related to their income, which means that willingness of education will be lower in the poorer family, but the willingness of human capital investment in richer family is weakly negative related to their income. Similarly, the dummy variable model and quadratic regression based upon the key variable also confirm the conclusions above. Secondly, contrary to the common sense, there’re not strong statistically significant relationships between the willingness of education and the regional economic development. There’re no significant differences between the rural areas and urban areas, and also no significant differences among different provinces. Thirdly, quantile regression results show an inversed-U shape relationship between willingness of high school education and household’s income, which means that the changes of household’s income have a strong impact on their willingness to accept or not to accept the human capital investment, especially for those with strongest or weakest willingness to accept human capital investment. For those with average willingness to accept human capital investment, the expectation of benefits and the opportunity cost from human capital investment are the main power to impact their willingness.In the fifth chapter, we focus on the issues that household vulnerability leads to poverty traps. Mostly poverty is transient for external shock, and this kind of poverty is not easy to identify, vulnerability is a key method to measure the recover ability of households under a shock. Based on the VEP method, we propose a reverse measure to calculate vulnerability, by which the vulnerability of a household is defined as the minimum level of welfare maintaining at a given probability of returning to poverty in the future. Based on a total of three years sample from CHNS survey in2000,2004and2006, we calculate households’ vulnerability at the level of0.1,0.3,0.5,0.7and0.9, and examine the characteristics of household vulnerability according to different household characters, such as the gender of the head of household, their locations and their occupations. The main results are summarized as follows. Firstly, there is a significant positive relationship between household income and their vulnerability and the vulnerability of high income household differences a lot as their income rising, which suggests that low-income households are always very vulnerable but high-income households sometimes vulnerable. Secondly, the degree of household vulnerability in different periods and regions is quite different. The overall vulnerability of the whole country are improving as time expans, the vulnerability of the household in eastern area are better than those in central and western area and the vulnerability of the households in urban area are better than those in rural area. Thirdly, the vulnerability of the household with different households’characters also varies. The vulnerability of male-headed households is slightly better than female-headed households, the households with more elders and children are more vulnerable than those with younger adults. Forthly, vulnerability of household with different occupation also varies. Households living by agriculture are the most vulnerable and the least vulnerable households are those living by senior skill. In order to investigate the impact of different factors on households vulnerability, we’ve also analyzed two types of factor decomposition problem, the within change and the between change. For the former, we selecte the deteriorating households in two periods of2004and2006, and then decompose their vulnerability into mean changes and variance changes. The analysis shows that the main cause for the vulnerability deteriorating is from changes both in mean and variance, and with equal magnitude. For the latter, according to a selected sample divided by the upper and lower1/3quantile line, by using the methods of Oaxaca decomposition, Shapley decomposition and Fields decomposition, results show that the burden information of a household, such as the household’s size, household’s expenditure and household’s structure are the main factors for households’vulnerability gap. Specifically, household expenditure is most important factor contributed to the vulnerability gap and its Shapley decomposition contribution is73.10%!In the sixth chapter, since empirical results indicate that individual income is closely related to their time preference and the rich are more patient than the poor, we consider an infinite and continuous time model, based upon the analysis of Meng(2006) to check how the impatience affects the economic stability, by postulating that individual agents’time preference depends on the economy-wide total income, average consumption and money balance, which are social factors taken as external by individual agents. Within extended neoclassical one-sector optimal growth model, our focus on the average social level is to check its effect on the macroeconomic stability, mainly using dynamic methods. We assume that labor supply is inelastic and the role that physical capital plays in stabilizing the real side of the economy by ensuring uniqueness of equilibrium in alternative environments if the monetary authority follows interest-rate feedback rules. The analysis shows that with the socially determined individual time preference, the macroeconomic stability of the long run equilibrium can arise under a set of sufficient conditions which is derived in this context.