发展经济学0928.pptVIP

第三章 新增长理论 经典增长模型的回顾 哈罗德—多马模型： 索洛模型： 稳态方程 外生技术进步下（Exogenous technical progress）的索洛模型： 3.1 收敛（Convergence）？ Absolute convergence (unconditional convergence): Lets assume that all the countries have the same steady state with requires that their s, n, δ, π and the production functions are the same. Poor countries grow faster than the rich ones. 无条件收敛图解 Empirical evidence The growth rates of homogenous countries do converge more clearly than the growth rates of non-homogenous countries (the US states, OECD vs. the world). Homogenous countries are more likely to have “the same” steady state. Evidence (rich countries) But not for the poor Whole sample, 1960-1999 No absolute convergence ? Neither in the postwar period ? Nor in modern growth era (since1820, Maddison) ? So all countries grow at the same rate? – Need careful empirical analysis – There is conditional convergence ? Countries differ in s and n ? Hence may converge to different steady states Conditional convergence A country will grow the faster the further away it is from its own steady state. A poor country may thus grow faster than a rich country if we consider their different steady states. Assumption: the steady states of the countries differ (their parameters differ). “Convergence clubs” 条件收敛的图解 3.2 几个需要回答的问题 1.为什么各国人均收入差距如此之大？或关键参数方面的差距为什么存在？ 2.技术进步如何发生的？是天上掉下的馅饼吗？ 3.穷国和富国间的资本回报率存在巨大差距吗？资本从富国流向了穷国吗？ 3.3 人力资本与经济增长 Human capital to Solow model Acknowledge that labor is more productive if it is educated and skilled. Will Solow model fit the data if we acknowledge the different education levels？ Barro (1991): testing for convergenceconditional on human capital Dependent variable: average growth in real GDP/capita over 1965-85. Independent variables —GDP per capita in 1960 —enrollment in primary and secondary schooling Results: coefficients on education positive and significant and on per capita GDP negative as Solow’s model would p