An Enterprise Map of Belgium

“An enterprise map of Belgium,1 a guide for industrial policy” uses a unique dataset of approx. 305,000 Belgian enterprises2 observed during the period 1997-2013. We study the evolution and dynamics of these firms and develop a framework for industrial policy. The framework can serve as a guide for policymakers to decide which industries to actively develop. Studying the firm data, we observe manufacturing remains an important pillar of our economy, though its share in private sector employment and (less outspoken) its share in added value is shrinking. Jobs are lost in the manufacturing industry and created in services industries. Within the services industry both knowledge intensive services (KIS) and less knowledge intensive services (LKIS) experience similar growth percentage-wise. In absolute numbers, however, the less knowledge intensive services showed an increase in employment of 3-to-1 compared to the increase in employment of the knowledge intensive services during 1997-2013. The increase in LKIS is to a large extend driven by the system of “service vouchers”, a system of government wage subsidies targeted at domestic services. On average, real remuneration per hour increased over the period, but the underlying re-allocation between industries shows a large heterogeneity with different effect on different groups: higher paying manufacturing jobs are predominantly replaced by lower paying services jobs. Also in Belgium, there is an ongoing trend of job-polarisation between higher wage jobs found in knowledge intensive services and (what is left of) manufacturing on the one hand and lower wage jobs found in less knowledge intensive services (distribution and service vouchers related sectors) on the other hand. Geography wise, we also find diverging trends. Whilst the less knowledge intensive services experience similar growth across counties (arrondissements), manufacturing and knowledge intensive services do not. This might also lead to a geographical polarisation between different areas. One potential driver of the these trends is further scrutinised, i.e. the incidence of high growth firms. We know that a limited number of high growth firms play an essential part in job creation and reallocation. We find a clear decline in the amount of high growth firms. This could lead to a loss of dynamism and entrepreneurship in our economy. Policy makers needn’t necessarily remain bystanders as these trends emerge. Recent evidence has shown that industrial policy can in fact result in higher economic performance, provided it is well designed. We develop a framework for an industrial policy based on the level of innovation and specialisation in different manufacturing industries. We find a significant potential for the pharma and transport & automotive industry. The textiles & apparel industry and electronics and electrical equipment manufacturing could be further developed into future key industries. They will need, however, increased levels of innovation and further specialisation.nrpages: 41status: publishe