New Research Article on Measuring Port Efficiency with Flexible Capital Inputs

In the new academic publication Mitigating Simultaneity Bias in Seaport Efficiency Measurement, Vista Analyse, the Institute of Transport Economics, and the University of Turku address estimation biases in the existing literature on port efficiency. The biases in conventional efficiency estimation stem from the simultaneous determination of port capital, production, and efficiency within the same period.

The research team behind the article includes Kenneth Løvold Rødseth from the Institute of Transport Economics, Timo Kuosmanen from the University of Turku, and Rasmus Bøgh Holmen from Vista Analyse. The article is published in Transportation Research Part A: Policy and Practice, a journal ranked at the second level in the Norwegian publication system, the highest tier in the system.

The article can be accessed here. Rasmus Bøgh Holmen may be contacted for any inquiries.

Correcting Estimation Errors

Since the 1960s, productivity estimation in conventional econometrics has focused on addressing biases related to the simultaneous determination of productivity and production inputs in the same period efficiency is realized. Around the turn of the millennium, a new approach emerged to tackle this issue, introducing a control function with a proxy for production shocks in the estimation (see, for example, Olley and Pakes 1996 and Levinsohn and Petrin 2003).

However, this issue has received less attention in the more specialized field of productivity estimation, often referred to as frontier estimation. Frontier estimation distinguishes between best practices and efficiency relative to those best practices. Traditionally, this literature has been dominated by two camps; Stochastic Frontier Analysis (SFA), which focuses on accounting for stochastic noise in regressions, and Data Envelopment Analysis (DEA), which avoids imposing strict assumptions about the functional form of cost and production functions. The relatively new estimation methodology Stochastic Nonparametric Envelopment of Data (StoNED) combines both approaches, accounting for stochastic noise in data, while allowing for varied functional forms of cost and production functions across contexts.

"In our contribution, we integrate the control function approach into the StoNED estimation method. Few studies in the frontier estimation literature have previously highlighted the biases associated with production inputs," Holmen explains.

A Maritime Context

Efficiency measurement for ports is one of the most popular research topics in maritime economics. In the empirical section of the article, the researchers examine Norway’s eight largest container ports, making parts of the work relevant for Norwegian port authorities.

This is the third article the researchers have co-authored on port efficiency. The first article, Market Access and Seaport Efficiency: The Case of Container Handling in Norway, published in Journal of Shipping and Trade, explored the importance of market access for efficiency measurement. The second article, Nonparametric Estimation of Allocative Efficiency Using Indirect Production Theory: Application to Container Ports in Norway, published in Journal of Productivity, addressed the difference between optimal and actual capital composition in Norwegian container ports.

"I want to thank Kenneth Løvold Rødseth and Timo Kuosmanen for an excellent collaboration on yet another research article. We look forward to continuing our work on productivity, capital, and maritime studies," Holmen concludes.