Comparison of Capital Costs per Route-Kilometre in Urban Rail

TL;DR

This paper compares capital costs per route-kilometre of urban rail projects, revealing high variability influenced by construction type, ground conditions, and other factors, highlighting the need for more data for robust analysis.

Contribution

It provides initial comparative insights into urban rail costs across projects, emphasizing the variability and factors influencing costs, and identifies the need for further research.

Findings

Costs mainly between US$50-100 million in Europe

US costs range from US$50-150 million

High variability due to multiple project-specific factors

Abstract

Because of the prominent position of urban rail in reducing urban transport-related problems, such as congestion and air pollution, insights into the costs of possible new urban rail projects is very relevant for those involved with cost estimations, policy makers, cost-benefit analysts, and other target groups. Knowledge of the differences in costs per kilometre, including explanations of differences and their breakdowns is currently lacking in the literature. This paper aims to provide a first stage insight into how cost per kilometre varies across urban rail projects. The methodology applied is a simple cost comparison across projects where the data collected are comparable. We conclude that capital costs per route-kilometre of urban rail vary highly between projects. Looking at European projects and excluding outliers, the total capital costs per route-kilometre (including stations and rolling stock) lie mainly between US$50-100 million (2002 prices). Including US projects, the range is US$50-150 million. The main reasons for the high variation in the route-kilometre costs are differences between projects as regards the ratio of underground to above-ground construction, ground conditions, station spacing, type of rolling stock, environmental and safety constraints and labour costs. We warn, however, that the observations used to reach the conclusions are too few to obtain results with statistical significance. Our results must therefore be seen as a first step towards collecting more data so that a more succinct statistical analysis can be conducted. Another conclusion is therefore that this area has future research potential.