Cost Comparisons for Metro Tunnelling Projects in 4No. Major World Cities.

J. F. Benson, N. R. Wightman

Sinclair Knight Merz (HK), Hong Kong SAR, China.

A. D. Mackay

Nishimatsu Construction Co. Ltd, Hong Kong SAR, China.



ABSTRACT: Over the past decade, extensive tunnel construction for underground Metro rail projects in major world cities has been undertaken with varying degrees of success. This paper analyses the tunnelling cost data from Metro projects in Hong Kong, Singapore, Shanghai, and London. The costs are compared for the key items of civil, architectural, building services, electrical and mechanical, permanent way, signalling / control and rolling stock. Specific tunnelling cost sensitive issues are discussed to help explain significant cost fluctuations. The conclusions highlight some of the influencing factors that contribute to more effective budgeting and forecasting of costs on underground Metro rail projects. Of significance is that low labour costs may not necessarily lead to lower tunnelling costs. Other key project elements show significant differences between different countries for a variety of reasons.


1. Introduction

Populations’ propagating towards cities and the need for efficient transportation has generated the need for better underground space utilisation and the development of Metro systems in major world cities since the 1950’s.

This paper examines the generic costs associated with tunnelling related projects and the reasons for variations in 4No. countries, and aims to provide insights into some of the financial aspects and cost differences associated with underground Metro systems in 4No. major cities within specific countries.

It is well known in the tunnelling industry that cost overruns and large financial claims on underground urban or tunnelling projects tend to be the norm in some countries. These cost overruns are often associated with the civil and underground elements of the projects. This paper touches on some of the reasons for cost variations between different countries for similar major underground infrastructure Metro projects.



Producing cost estimates for tunnelling and Metro projects is very complex because every project and the ground conditions are always different. There are many variables that influence costs, and generally the best method to prepare a cost estimate from first principles, requires knowledge of the ground conditions, risks, existing constraints, basis of design, feasible construction methods and programme. Furthermore, in every country there can be different ventilation and fire and life safety strategies, specifications, construction standards, labour laws and expectations with respect to safety, quality, environmental issues and risk management.

In the paper ‘Comparison of Capital Costs per Route-Kilometre in Urban Rail’ (Flyvbjerg B. et al, 2008), the basis of comparison for Metro projects is linear kilometre. As opposed to a linear metre comparison, the authors of this paper consider that a better comparison is achieved by comparing the cubic metre of excavated and constructed Metro tunnels, stations, shafts and other infrastructure. It is noted that a cost comparison based on cubic metres is still relatively crude, but it is suggested that this does provide a reasonable high level or ‘ball-park’ cost estimate for comparison.

Published tender data, (China Investment Consulting Net, 2011; MTRCL, 2010; LTA, 2010 and Montague, A, 2003.), has been gathered from various sources and a basic knowledge of the station, entrance, shaft and tunnel sizes has been used to analyse and generate ‘like for like’ cost comparisons. It is accepted that this is very approximate.


In order to establish a benchmark for cost comparison, a Metro project, namely the West Island Line in Hong Kong has been used as the base case. This project involved a wide combination of construction methods that could be used to form the basis of comparison on a cubic metre of excavation basis.

The methods of construction included:

  • Cut and Cover

  • Drill and Blast (Mined Tunnels)

  • Soft Ground Tunnelling (SCL)

  • Ground Treatment (Micro-fine grouting)

  • Ground Treatment (Ground Freezing)

  • Tunnel Boring Machine (TBM)

Special consideration was given to major cost items such as TBMs, obstructions and ground treatment that could distort comparisons. These items were separated to produce a more accurate cost comparison. Further breakdown and consideration was also given to variations in construction techniques, to establish further refinement of the cost comparisons.

For example, with cut and cover construction, further breakdown was carried out to compare ‘bottom-up’ and ‘top-down’ construction, and specific techniques such as silent piling, secant piling, pipe piling and diaphragm walls.


3. Data Comparisons

The data has been gathered from Metro projects that have been tendered or constructed at different times albeit within a similar period between the years 2000 and 2013. The data has been gathered from a variety of sources as noted in the references.

Table 1 below is based on information gathered from the Hong Kong West Island Line (WIL) project, which consisted of 1 No. Cut and cover station (Kennedy Town) and 2 No. Cavern stations (University and Sai Ying Pun) and 3.3km of rail tunnels constructed that connect onto the already operating Island Line at the existing Sheung Wan Station.

It is important to note that a large percentage of this project was constructed in granite rock, and the soft ground tunnelling required a state of the art TBM, along with extensive ground treatment, instrumentation and monitoring.


Table 1. Hong Kong – MTRCL – West Island Line (Baseline) – 2006 and 2010















































The different construction methods required a detailed breakdown and analysis in order to compare the costs associated with the WIL project, with other Metro projects, in different ground conditions, in other countries.

On Table 1, the column on the right indicates the price difference or fluctuation between the 2006 budget estimate and the actual 2010 tender prices.


3.1 - Hong Kong – MTRCL - West Island Line (Benchmark) – 2006 and 2010

It is noted that even though the Hong Kong WIL prices increased significantly from 2006 to 2010 due to inflation and escalation, the original budget estimates were proportionally close to the actual tender prices (adjusted for inflation), with the exception of the Civil works. It is noted that the increase in Civil costs was 2.89 times (2006 to 2010), and was due to Hong Kong being in a mini-recession from 2003 to 2007 and then construction worker salaries increasing 10% per year between 2006 and 2010 as the construction industry in Hong Kong shifted from a ‘bust’ cycle to ‘boom’ when 4 No. major Metro tunnelling projects were released. It is also noted that % split increased for civil works by 7.2%.


3.2 - Cost comparisons with other metros in other cities

The generic costs of underground/Metro projects in Singapore, Shanghai and London were then compared against the Hong Kong WIL project, which was used as the cost benchmark, in Table 2.

The cost comparison process was based on cubic metres, and endeavored to normalise the costs so that similar construction methods in similar ground conditions were compared.

This required the data to be broken down per cubic metre for the various construction techniques.


Table 2. Metro cost comparison between Hong Kong, Singapore, Shanghai and London. – Percentage Cost Difference relative to Hong Kong WIL in 2010

































4. Discussion on Cost differences between different disciplines and countries

4.1 - Cost differences between disciplines

The results of the analysis based on a per cubic metre basis revealed some interesting findings when compared to the Hong Kong WIL project, and some of the more significant differences are discussed below. 

In general the cost differences between Hong Kong and Singapore are similar but there were two significant cost differences, namely the civil costs, and contingencies or claims. In Singapore, the ground conditions consist of some of the hardest rocks (Bukit Timah granite) and softest clays (Kallang Formation). These variations influence the construction costs to generate higher Civil costs in the order of 12.7%. However, in Singapore, there is another influencing factor that taints underground construction costs, and that is the lowest price tender philosophy, where the reluctance to lodge and then generate successful claims results in the Contingencies and Claims being 6.1% lower. Labour is imported in Singapore for much of the construction work, but minimum safety standards are still required which increases staff training costs.

It was a surprise to find that the electrical, mechanical and permanent way costs in Shanghai were 9.2% higher, and the signaling and control systems were 7% higher. This was explained as being due to using non-standard equipment. As China has become more of an international power and as foreign investment and technology has been introduced, we notice that these costs are significantly lower.

It was also a surprise to find that in Shanghai, the rolling stock was 8.4% higher in 2012. This was unusual as Chinese rolling stock supply orders have become very popular in a number of European and Asian countries in 2013 (Malaysia, Hong Kong and Argentina).

We notice that China’s determination to dispel corrupt practices, with the high profile sacking of industry leaders has recently influenced this sector.

In Shanghai the contingencies and claims were 9.5% lower. Since the projects were government funded and constructed by state owned companies this is not unusual.

In Shanghai we found that client management costs were 9.5% lower. These lower costs are suggested to relate to poor planning and issues associated with the patronage modelling not matching the size of stations and distances between stations. The result in China is that a lot of the Metro stations are very large and close together. This was also found to be deliberate in some instances or the result of one Chinese city copying another Chinese city’s metro designs.

It is also noted that the early metros in China were constructed in the 1950’s when there were concerns with the China-Russia relationship. At the time, this led to an urgent need to have safe wartime refuges below ground; like London’s underground stations, which served it well in the Second World War.

In London, between 2008 and 2012, a significant downturn was experienced by the UK economy. This affected labour costs in the tunnelling and Metro markets with limited underground construction work until CrossRail proceeded.  This helps to explain why a large contingent of UK tunnelling resources shifted to Hong Kong during this period and why the Civil works costs are 4.2% lower in London than compared to Hong Kong.

In London, the signalling and control costs were estimated to be 9.3% higher on CrossRail due to complexities associated with aged existing systems and infrastructure.

In London, it was noted that costs associated with contingencies were 6.1% lower, and client Management was 5.5% lower. It is worth noting that the effects of potential damage to existing buildings is well understood in London, as a result of extensive instrumentation and monitoring on the Jubilee Line extension and Heathrow Express tunnel projects. Also, the costs associated with building protection tend to be included with land issues as opposed to contingencies and client management. This helps to explain some of the higher land cost issues in London as well as the lower contingencies and client management costs. In London the land issue costs were 9.0% higher. There are a variety of reasons for this, with the main issue being a long drawn out protracted land acquisition processes, as well as the offsetting effects from the contingency and client management costs noted above.  

Compensation claims contribute significantly to the higher costs associated with land in London, and often there are developers and landowners who strive to achieve the best commercial advantage possible from Metro and underground projects.

In addition, in the UK there is also a robust public consultation process and a strong demand for stations to be located at different locations, which can generate costly litigation.  Somewhat unique to London, are the underground English Heritage archaeological discoveries, which if encountered, can have significant impact on both programme and costs.


4.2 - Cost differences between disciplines

The management philosophy differs between countries and it can be difficult to define; often encompassing risk management, safety and anti-bribery and corruption.

In countries like the UK, Singapore and Hong Kong the markets tend to be mature and well developed. There are often government departments, or statutory regulations that significantly influence costs. In the United Kingdom, there are Construction Design Management (CDM) regulations; in Hong Kong, the Buildings Department (BD) and in Singapore the Building and Construction Authority (BCA).

In mature markets there are often, also institutions that share knowledge and collectively improve standards or resolve issues for the betterment of society. In mature markets, independent checking and reviewing is usually accepted as a necessary process to maintain standards. Conversely, mature markets can be prone to an excessive high degree of statutory process or ‘red tape’; or the ‘interpretation’ of such requirements, which can result in delays and cost implications.

As noted in Table 1, over 50 percent of the cost of an underground Metro project is associated with the civil costs and ground conditions. Based on the findings of the study, Tables 3 and 4 below provides further explanation on cost differences between countries.


Table 3. Primary reasons for underground metro cost differences between specific cities.
















































Table 4. Secondary reasons for underground metro cost differences between specific cities.
















































5. Conclusions and recommendations

It has been demonstrated that comparing the costs associated with underground construction of Metros systems in 4No. different countries are very complex due to the many parameters that affect the total costs. The main findings revealed that lower labour costs may not necessarily lead to lower tunnelling costs

This paper does provide some explanations surrounding the cost differences associated with underground Metro systems in Hong Kong, relative to Singapore, Shanghai and London, along with the primary reasons affecting the magnitude of the cost fluctuations.

It is also noted that the data pool is very limited and this study would benefit from using further information from Metro projects in other world cities across the globe. Also, when the final payments and settlements are made on the projects studied, there may be further information and explanations to share.

The findings in this paper indicate that there are a wide variety of reasons for cost differences that are not always related to labour, plant and material costs.

Further research into effective risk management techniques versus procurement methods, relative to the cost differences would also provide further insights into the effects that procurement options have on cost estimates.




The authors wish to thank their many friends in the global tunnelling industry who have provided discrete information to substantiate and verify the comments in this paper. These sources remain confidential.


China Investment Consulting Net. 2011. Analysis and Investment Consulting Report on China Metro Industry, 2011-2015.

MTRCL. 2010. Published tender award data. Hong Kong Mass Transit Rail Corporation Limited  

LTA. 2010. Published tender award data. Singapore Land Transport Authority.

Montague, A. 2003. Published tender award data. Crossrail Review 2003-2004, UK.

Flyvbjerg, B,; Bruzelius, N. and Van Wee, B. 2008. Comparison of Capital Costs per Route-Kilometre. Urban Rail. EJTIR, v.8, no.1. p 17-30.




© 2014 by Benson Consulting (HK) Ltd. Proudly created by