Answers are grounded in this paper and cite their sources.
Type of the article: Research ArticleAbstractResearch on maritime clusters and project-based production networks often focuses on aggregate crisis responses or individual disruptions. Less is known about how relative efficiency differs across firm types and market regimes over prolonged periods. This study examines equipment providers, large yards, and small and medium-sized yards in the Møre og Romsdal maritime cluster and production network from 2010 to 2023, covering the offshore boom, oil-price collapse, recovery, COVID-19 pandemic, and post-pandemic period.Efficiency is estimated using input-oriented, variable-returns-to-scale Data Envelopment Analysis (DEA) based on accounting data. Two random-effects Tobit models examine temporal and structural associations in the DEA scores, supported by firm-clustered bootstrap inference, predicted margins, average marginal effects, and fractional logit robustness tests.Results show persistent, regime-varying differences across firm types. Equipment providers maintain the most stable relative position, large yards show the greatest variation, and small and medium-sized yards occupy a comparatively weaker position for much of the period before improving in the post-pandemic period. Large yards reach their weakest predicted relative position during recovery rather than during the immediate oil-price collapse. Regression results further show a positive association with monetary conditions, a negative association with petroleum-investment conditions, and a specific large-yard disadvantage during recovery, while liquidity and cluster-level production dispersion are less stable across specifications.The findings are consistent with a structural-contingency perspective; different firm configurations are associated with distinct relative efficiency patterns as environmental conditions change. External disruptions should therefore not be treated as homogeneous events with uniform implications across the maritime production network.
No takes yet. Share an insight, caveat, or question.
Danielsen et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: