Study on Economy-wide Impact Assessment of Public Procurement for Innovation for New Emerging Industry: A Computable General Equilibrium Approach

Abstract

Recently, there is growing interest in demand-side innovation policy to guarantee enough demand for innovative suppliers. Public procurement for innovation (PPI) is one of the representative demand-side innovation policies, and has been evaluated as the most efficient policy tool for stimulating innovation. Many countries allocate a large proportion of government spending to public procurement, and most public demand in some technologies or industries, such as energy, environment, healthcare, and construction, is generated by public procurement. Despite the interest and effectiveness of public procurement for innovation, there is little work on policy impact assessment of public procurement for innovation from an economy-wide perspective. Most previous research has been limited to case studies and qualitative analysis. However, since public procurement for innovation has not only a direct demand effect but also indirect effects related to production, private demand, and innovation, it is worthwhile to analyze the policy impact of public procurement for innovation with an integrational approach. Therefore, this study analyzes the various paths of policy impact of public procurement for innovation with a literature review and investigates the economy-wide effect of public procurement for innovation with knowledge-based computable general equilibrium (CGE). This study chooses the electric vehicle (EV) industry to conduct empirical simulation and modified knowledge-based CGE model with respect to public procurement and learning effect. Thus, when there is a low R&D subsidy and a high sales incentive for EV with PPI, the policies achieve both EV market expansion and economic growth simultaneously. In other words, not all policy mixes achieve the policy goals of both GDP growth and EV industry expansion. This is because the policies have a reverse effect on other industries that are not targeted by the policies when the policy mix is different. Because there is little generation of private production of other industries and the government expenditure is used for PPI, the public and private R&D are not much higher than they would be if there were no policies. Finally, because of the increase in sales volume of EV, the EV industry experiences learning by doing and economies of scale, which reduces the requirement for primary factors on the production process. This study addresses the question whether economy-wide policy impact assessment is necessary before policy implementation, especially for innovation and industrial policy in certain sectors. The integrational framework suggested in this study could generate numerical and quantitative evidence for ex-ante policy impact evaluation, and it could be utilized for other innovation and industrial policy and policy mixing.