The Dynamics of Innovation, Competitiveness and EmploymentPanel Evidence for Luxembourg


CALL: 2012

DOMAIN: SC - Social and Economic Cohesion

FIRST NAME: Wladimir

LAST NAME: Raymond




KEYWORDS: Dynamics, Innovation, Competitiveness, Employment, Profitability, Markov Chains, Panel Data, Simultaneous-Equations, Nonlinear Models

START: 2014-01-01

END: 2016-12-31


Submitted Abstract

The project seeks to investigate empirically the extent to which innovation performance by Luxembourgish firms affects their competitiveness and their ability to create jobs. The dynamics of these three components will also be studied. More specifically, we are interested in the following research questions.• To what extent does technological innovation performance by the firm, e.g. innovative sales and unit cost reduction resulting from product and process innovations, affect its competitiveness and its ability to create jobs? • Are there innovation strategies, i.e. combinations of technological and non-technological innovations, which result in greater competitiveness and job creation? (Henceforth, we will refer to these strategies as ‘winning strategies’.) Are technological and non-technological innovations complements or substitutes?• If such ‘winning strategies’ exist, how likely is the firm to keep them in the future? In other words, are these strategies persistent or transient?• Similarly, are competitiveness and job creation resulting from ‘winning strategies’ persistent or transient?We will tackle these questions by using nonlinear dynamic simultaneous equations models using firm-level panel data during the period 2002-2010.The data that will be used stem from several waves of the Luxembourgish Community Innovation Survey (CIS) supplemented with data from the Structural Business Statistics (SBS) and the Investment Survey (IS). The Luxembourgish CIS provides several qualitative and quantitative measures of innovation. Among the qualitative measures are binary variables indicating whether a firm is a product, process or an organizational innovator. Combinations of these binary variables form the above-mentioned innovation strategies. Among the quantitative measures are the share of innovative sales (sales of new or significantly-improved products over total sales) and the reduction in unit cost (in percentage) due to process innovation. A sufficient measure of firm competitiveness is profitability captured, for instance, by return on investment, return on assets or profit margin. These variables are readily available or can be computed in the SBS. Employment figures are readily available in both the CIS and the SBS. In order to express the firm’s ability to be competitive and to create jobs, growth rates of profit and employment will be considered. As for the methodology, regardless of the theoretical or conceptual framework considered, nonlinear dynamic simultaneous equations models will be estimated. The structural models will consist of three blocks, each of which uncovers the dynamics of innovation, competitiveness and employment. The estimation of such models poses formidable challenges. One of them is controlling for dynamics and individual effects. In order to achieve so, we will need to render existing estimation techniques, e.g. full information maximum likelihood (FIML), suitable for our framework or devise new estimation techniques. In both cases, we intend to carry out Monte Carlo experiments to assess the accuracy of these techniques when applied to our case.

This site uses cookies. By continuing to use this site, you agree to the use of cookies for analytics purposes. Find out more in our Privacy Statement