Technology Criticality, Advanced Material, Firms' Longevity
Description
This dataset contains comprehensive survival, structural, and technological information on 17,997 exited firms operating in the Advanced Materials Technology (AMT) sector across 58 countries. It is designed to examine how institutional conditions and technological environments shape the longevity of firms engaged in chemicals, plastics, specialty materials, pharmaceuticals, coatings, and related subsectors classified under the NACE industrial system. The unit of analysis is the “mortal” firm—a firm that exited the industry and had at least five employees at the time of exit. Firm longevity is measured as the number of days (or years) between establishment and confirmed exit, with right-censoring applied for firms still active at the end of the observation period. The dataset spans firms established between 1971 and 2019, offering multi-decade survival patterns within technology-intensive industries. Three core predictors structure the dataset: value-chain position, legal structural complexity, and technology criticality. Value-chain position is coded from upstream to downstream stages, while legal structural complexity ranges from sole proprietorships to fully incorporated public limited companies. Technology criticality is represented through a weighted 1–10 scale capturing economic importance, supply-chain risk, strategic value, regulatory implications, and substitutability of material technologies. The dataset also includes extensive firm-, industry-, and location-level controls, such as group size, industry competition, consolidation status, ownership dispersion, revenue, employment, OECD membership, post-2007 crisis indicator, and city-level mortality ratios. These enrich the analytical capacity of the data for modelling institutional, competitive, and technological pressures on firm exit. The dataset supports advanced statistical modelling, particularly Accelerated Failure Time (AFT) survival analysis, and is suitable for exploring nonlinear, interaction, and contingency effects. It offers valuable insights for scholars of industrial dynamics, strategic management, innovation systems, and policy makers concerned with the resilience, competitiveness, and structural vulnerabilities of firms in critical material technologies.
Files
Steps to reproduce
Steps to Reproduce the Dataset The following steps describe how to reconstruct the full dataset of 17,997 exited firms in the Advanced Materials Technology (AMT) sector, including value-chain position, legal structure, technology criticality, and all control variables. Step 1 — Define the Population and Industry Scope Use the NACE Rev. 2 classification system to identify AMT-related industries. Include all codes listed in Appendix 1 (e.g., chemicals, plastics, specialty materials, pharmaceuticals, coatings, adhesives, fibres, gases). Extract all firms established between 1971 and 2019 within these NACE classes. Step 2 — Extract Firm-Level Records Acquire firm-level administrative data from business registries, commercial databases (e.g., Bureau van Dijk / Orbis), or national statistical offices. Retrieve the following attributes for each firm: Establishment date Exit/deregistration date Legal form Revenue Employees Ownership structure Group affiliation Consolidation status Step 3 — Determine Firm Longevity Compute longevity as: Longevity = Exit Date − Establishment Date Longevity=Exit Date−Establishment Date For firms active at the observation cutoff, mark them as right-censored. Step 4 — Construct the Value-Chain Variable Map each NACE class to one of six value-chain stages (upstream → downstream). Assign codes 1–6, following Gereffi (2019), Porter (1990), and empirical mappings. Step 5 — Code Legal Structural Complexity Convert national legal forms into the six-level hierarchy in Appendix 2: 1 = Sole proprietorship → 6 = Public Limited Company Validate cross-country equivalences of legal categories. Step 6 — Compute Technology Criticality Scores Assign AMT technologies to one of 10 criticality levels based on: Economic importance Supply risk Strategic/national security value Substitutability Regulatory/safety concerns Apply the weighting model from Graedel & Nuss (2014) and Paulsen et al. (2018): Criticality Score = ∑ ( 𝐸 𝑖 × 𝑊 𝑖 ) Criticality Score=∑(E i ×W i ) Step 7 — Add Control Variables Extract or construct the following: Industry competition (number of firms in the same technology field) Group size Subsidiary vs. parent indicator Revenue (log) Employees (log) OECD membership Post-2007 crisis dummy City mortality rate (city exits / national exits) Step 8 — Clean, Transform, and Filter the Dataset Remove firms with missing revenue (as in the original study). Log-transform skewed variables (revenue, employees, competition). Ensure consistent country IDs and time formats. Retain only firms with ≥5 employees at exit. Step 9 — Final Quality Checks Verify longevity distributions (mean ≈ 7200 days). Confirm final sample size (N = 17,997). Validate ranges of key variables (criticality 1–10, legal structure 1–6, value chain 1–6).
Institutions
- Liaoning University