Regional competitiveness and high growth firms in the EU: the creativity premium

ABSTRACT

The finding that the economic performance of regions is driven by a small set of productive firms calls for a granular analysis of regional growth and competitiveness. Among the productive firms, a significant contribution is made by so-called high-growth firms (HGFs), which also account for the bulk of new jobs created in the region. In this paper, we examine the variation in the incidence of HGFs across a wide set of EU regions in relation to the regional entrepreneurial ecosystem. Looking through the lens of the experimentally oriented economy, we conceptualize the essential role that the creative class plays in the formation of collaborative innovation blocs from which dynamic entrepreneurship and high-growth firms emerge. We present original evidence showing that, for the human capital present in the region, employment in creative occupations has the strongest impact on the occurrence of high-growth firms. Our evidence also points to the stimulating role played by the quality of market-supporting institutions, agglomeration, and infrastructure.

KEYWORDS:

• High-growth firms
• creative class
• entrepreneurship
• regional development
• regional policy

JEL CLASSIFICATION:

• R11
• R12
• M13

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ LASSO is a penalized regression method that minimizes the residual sum of squares subject to a constraint or penalization placed on absolute coefficient size. In our analysis, lambda, the degree of penalization, was obtained through 10-fold cross-validation, meaning the sample was randomly divided into 10 bins with each bin repeatedly becoming the validation dataset and the remaining 9 bins the training dataset. The model (i.e. the relation between HGFI and RCI variables) was then fitted to the training data, the prediction error calculated from the validation data, and the value of lambda ultimately identified as that which optimized predictive performance.

$RS{S}^{LASSO}=\sum _{i=1}^n{\left(y_i-{\mathit{x}}_{\mathit{i}}\mathit{\beta }\right)}^2+\lambda \sum _{j=1}^p\left|{\beta }_j\right|$.

² NUTS 2 stands for Nomenclature of Units for Territorial Statistics. Level 2 corresponds to regions varying in size between 800,000 to 3,000,000 people.

³ This variable is included in the creativity measure because the EU ‘creative class’ measure leaves out technicians in high-tech occupations and (management) workers in knowledge-intensive service industries, a category that is often included in the measures used in North-American studies to depict the creative class (see Falk et al. 2011).

⁴ Defined in the 2010 RCI as ‘employment in the financial intermediation, real estate, renting and business activities sectors (NACE rev. 1.1 J-K) as % of total employment’, a variable part of the ‘business sophistication’ sub-index.

Additional information

Funding

The financial support from Flanders Innovation and Entrepreneurship is gratefully acknowledged.