Technical Efficiency of Romanian Danube Ports. A Stochastic Frontier Analysis

Abstract

This paper evaluates the technical efficiency of Romanian inland ports along the Danube River over the period 2020–2024 using a Stochastic Frontier Analysis (SFA) framework. A panel data model with time-invariant inefficiency is estimated based on a Cobb–Douglas production function, where total cargo throughput is the output and quay length, number of quay cranes, and storage area are specified as inputs. The analysis covers 16 ports administered under the Upper and Lower Danube authorities. Results indicate a low average technical efficiency and substantial heterogeneity across ports. Lower Danube ports generally outperform Upper Danube ports, although the difference is not statistically significant. The estimated production elasticities reveal decreasing returns to scale, suggesting that most ports operate beyond their optimal size. Cargo diversification shows no statistically significant association with technical efficiency. The findings challenge the assumption that larger or more diversified ports are inherently more efficient and support a shift toward efficiency-oriented port development policies. The results should be interpreted considering the relatively short five-year available panel, which captures recent operational conditions, including the market disruptions associated with the Russian invasion of Ukraine.