Authors: Davide Castellani, Silvia Cerisola, Giulia Felice, Emanuele Forlani and Veronica Lupi

This report describes the computability and the availability of a set of competitiveness indicators calculated through a bottom-up approach, which consists in the computation of a competitiveness’ index by aggregating firm-level data for a specific group of firms. For example, aggregation can consist in the computation of averages (but also variances or other moments of the distribution) across all firms in a given sector (or region), or across subsets of firms (such as exporters or multinational firms). For each index, three levels of aggregation are considered: country, sector, and region. The mapping of micro-level databases includes information on firms’ industrial sector (usually NACE Rev. 1.1 or Rev. 2), and geographical location (usually NUTS2 region). The mapping allows users to know if a given competitiveness index is computable at sector, and/or regional level for each country.

This report analyses computability and accessibility for 45 bottom-up indicators, which can be grouped into 4 conceptual areas: (i) productivity, (iii) Firms’ dynamics, (iv) International activities, and (v) R&D and ownership.

Due to some incomplete replies or lack of responses from some of our contacts at the National Statistical Institutes or National Central Banks, this report presents a mapping of existing firm-level databases that can be employed to calculate these indices in 15 EU countries. These countries are: Belgium, Bulgaria, Czech Republic, Estonia, Finland, France, Germany, Hungary, Ireland, Italy, Latvia, Portugal, Slovakia, Slovenia, and Sweden

Our main findings can be summarized as follows:

  • indicators for labour productivity, TFP, and international activities have the highest degree of computability, given that they require the use of basic items from balance sheet data and trade statistics, respectively.
  • if an indicator of efficiency is computable by export status, usually it is computable also by import status.
  • in the case of Bulgaria, Hungary, and Latvia, it is more difficult to retrieve information on firms’ ownership.
  • Belgium, Estonia, Finland, France, Hungary, Ireland, Italy, and Sweden offer the widest degree of computability.
  • Finally, also the indicators on firm’s dynamics and other activities present a relative widespread computability among the countries that we were able to consider at this stage.

Conversely, the access to micro-level databases is not an easy task for researchers, because of the confidentiality restrictions, or the unreported rules of access (number 9 in the summary tables). In some cases, the access is guaranteed to researchers under certain conditions and a submission of research proposal (e.g., France, Italy, Sweden).Germany provides an interesting case of improving the comparability of different micro level database, with the objective to generate a unique business micro dataset (KombiFid project).