Wearing a Red Hat ¨C The Impact of Activist Industrial Policy on Software Development in China
Description
The idea that the government should steer economic development by strategically hand-picking and managing certain industries is controversial but appeals to many developing countries that are eager to upgrade their industries. In this paper, I study China's recent quest to jump start the country's software industry and provide the first systematic evidence on the impact of activist industrial policy on the country's software development efforts. I assemble a unique dataset comprising all software development projects completed between 2000 and 2004. To illustrate the competitive dynamics under the influence of political economy, I borrow insights and methods from the software engineering literature and supplement the data with key technology variables derived from an expert panel survey and a semantic analysis of the detailed project descriptions and assessment reports.
Focusing on the idea sources, I classify all projects into two groups, one that is initiated by the various government science and technology programs and the other that is initiated by software producers themselves. I use propensity score matching method to address the selection problem that the former group of projects is intrinsically likely to be different from the latter and therefore a direct comparison of the two is flawed.I find that government initiated projects are 4% to 14% less likely to be adopted either internally or by the market than the projects initiated by software producers. When the government initiated projects are subsequently undertaken by the State Owned Enterprise producers (SOEs), it takes 6 to 7 months longer for the projects to be completed. However, when software companies contribute their own resources to government initiated projects, the adverse impact of government initiation on both project success likelihood and development efficiency is alleviated. The results are robust to different samples and estimation routines. In addition, my findings on the characteristics of the government initiated projects corroborate theoretical motivations for government intervention and at the same time highlight the concerns for successful implementation of industrial policy.