Learning, innovation and economic growth: A long-run model of industrial dynamics.
Learning, innovation and economic growth: A long-run model of industrial dynamics.
复制标题
学习、创新和经济增长：工业动态的长期模型。
DOI：
10.1093/icc/3.1.199
复制
发表时间：
1994
影响因子：
6.4
通讯作者：
B. Verspagen
中科院分区：
文献类型：
--
作者：
G. Silverberg;B. Verspagen
企业理论一直被一种新古典主义理论所主导，这种新古典主义理论从管理者熟悉的大多数现实生活特征中抽象出来。然而，经济往往可以通过特定的行为规则来描述，无论是随着时间的推移，还是跨部门的。拒绝将完全优化的行为作为对经济活动的解释，并没有挑出任何精确的替代方案作为一种有限理性行为的理论。本文按照“人工世界”的建模哲学，建立了一个演化模型，描述了内生技术变革与经济增长之间的关系。在这个模型中，公司必须确定他们的投资预算中将有多大比例投入到研发中，作为一种日常运营方式。与更多的新古典方法不同，本文在有限理性的背景下研究了这一决策问题，在有限理性的背景下，代理人对他们的行为和结果之间的关系只能有模糊的想法。该模型围绕三个基本模块构建。第一块描述了人工经济是如何随着一组给定的技术和公司而演变的，选择发生在两个层面上。这一块由资本积累速度、新技术在企业总资本存量中的扩散和实际工资率等方程式组成。第二块描述了一套用于将新技术和公司引入经济的规则。这一块把企业的创新行为作为给定的，然后描述单个企业进行创新的概率以及这种创新是如何引入的。第三部分描述了创新行为如何在经济发展和企业学习的影响下发生变化。这一块描述了从表现到创新行为的反馈，从而是集体学习的一种形式。这篇论文的结论是，进化论方法似乎为一个经济体如何通过一系列增长阶段和市场结构在历史时期“自给自足”提供了一个有吸引力的替代解释。该模型还表明，对企业理论的有限理性方法，加上分析市场选择和集体行为的进化框架，确实产生了红利，这既可以解释可识别的行为模式如何出现在追求利润的假设中，而不是完全理性的利润最大化假设，也可以解释市场结构和增长机制如何同时内生。
The theory of the firm has been dominated by a neoclassical apparatus that abstracts from most real–life features familiar to managers. Yet the economy can often be characterized by certain regularities of behavior, both over time and cross-sectionally. Rejection of fully optimizing behavior as an explanation of economic activity does not single out any precise alternative as a theory of boundedly rational behavior. This paper develops an evolutionary model describing the relation between endogenous technological change and economic growth along the lines of an ‘artificial world’ modeling philosophy. In this model firms must determine what proportion of their investment budget will be devoted to R&D as an operating routine. In contrast to themore neoclassical approaches the paper investigates this decision problem in the context of bounded rationality where agents can have only vague ideas about the relationship between their actions and outcomes. The model is constructed around three basic blocks. The first block describes how the artificial economy evolves with a given set of technologies and firms, with selection taking place at both levels. This block consists of equations for the rate of capital accumulation, the diffusion of new technologies in the total capital stock of the firms and the real wage rate. The second block describes a set of rules that is used to introduce new technologies and firms into the economy. This block takes the innovative behavior of firms as given and then describes the probability that individual firms will make an innovation as well as how this innovation is introduced. The third block describes how innovative behavior changes under the influence of the evolution of the economy and firm learning. This block describes feedback from performance to innovative behavior and thus a form of collective learning. The paper concludes that an evolutionary approach appears to offer an attractive alternative explanation of how an economy can ‘bootstrap’ itself in histrical time through a succession of growth phases and market structures. The model also demonstrates that a bounded rationality approach to the theory of the firm, coupled with an evolutionary framework for analyzing market selection and collective behavior, does yield dividends both in terms of explaining how identifiable patterns of behavior emerge from profit-seeking rather than completely rational profit-maximizing assumptions and how market structures and growth regimes may be endogenized simultaneously.