Competing R&D Strategies in an Evolutionary Industry Model

Murat Yildizoglu

No 343, Computing in Economics and Finance 1999 from Society for Computational Economics

Abstract: Early evolutionary models of industry dynamics have used very simple ways of modeling bounded rationality. In the precursory work of Nelson and Winter (1982), for example, R&D decisions of firms are given by a fixed rule: firms invest in each period a fixed proportion of their capital stock in imitative and innovative R&D. Recent models have introduced more elaborate ways of modeling learning with bounded rationality, implicitly through replicator dynamics or simple adaptive mechanisms or explicitly through genetic algorithms or classifiers. Oltra & Yildizoglu (1998) provides a thorough analysis of different alternatives and proposes a general approach. In this work, I adopt a simpler framework to study the role of learning in industry dynamics. I use a simplified version of the initial model of Nelson and Winter (1982) that aims to neutralize the effects of the very peculiar capital-investment decision used in this model. With this version and its well-specified dynamics, I study the confrontation of two different types of investment behavior in Research and Development. The first corresponds to an updated version of Nelson and Winter's fixed-rule behavior: in each period, each firm invests a fixed proportion of its cash-flow on R&D. The second type of behavior includes learning: firms try to adapt their R&D/Cash-Flow ratio to the conditions of the industry. Learning is modeled here through the use of genetic algorithms by this type of firm. Both types of firms coexist initially in the industry. This simple framework is used to answer several questions that can be grouped under two headings: 1) The use of fixed R&D rules does not contradict the empirical evidence. One effectively observes quite stable R&D/CF ratios in industries, but it is important to study if this type of behavior is coherent with the presence of learning or if it can be endogenously generated in evolutionary models. 2) More theoretically, it is important to see if the explicit inclusion of learning in industry models is worthwhile: Does it enrich our understanding of technology dynamics? Does it suggest a competitive edge for strategies strongly based on learning? Does learning give a better chance of success in the long term? These questions are studied in a simulation program developed in Java. A first version of the program is already available in my web site.

Date: 1999-03-01
New Economics Papers: this item is included in nep-cmp, nep-evo, nep-fin and nep-ind
References: Add references at CitEc
Citations: View citations in EconPapers (7)

Downloads: (external link)
http://cournot.u-strasbg.fr/yildi/files/learnind.pdf main text (application/pdf)
Our link check indicates that this URL is bad, the error code is: 500 Can't connect to cournot.u-strasbg.fr:80 (A connection attempt failed because the connected party did not properly respond after a period of time, or established connection failed because connected host has failed to respond.)

Related works:
Journal Article: Competing R&D Strategies in an Evolutionary Industry Model (2002)
Working Paper: Competing R&D Strategies in an Evolutionary Industry Model (2002)
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:sce:scecf9:343

Access Statistics for this paper

More papers in Computing in Economics and Finance 1999 from Society for Computational Economics CEF99, Boston College, Department of Economics, Chestnut Hill MA 02467 USA. Contact information at EDIRC.
Bibliographic data for series maintained by Christopher F. Baum ().