This article was downloaded by: [University of Toronto Libraries]

On: 10 August 2014, At: 14:44

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

Building Research & Information

Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/rbri20

Construction costs comparison between ‘green’ and

conventional office buildings
a b
Michael Rehm & Rochelle Ade
a
Department of Property , School of Business and Economics, University of Auckland , 12
Grafton Road, Auckland , 1142 , New Zealand
b
Ade Consultants , 119 First View Avenue, Beachlands, Auckland 2018 , New Zealand E-
mail:
Published online: 07 Mar 2013.

To cite this article: Michael Rehm & Rochelle Ade (2013) Construction costs comparison between ‘green’ and conventional
office buildings, Building Research & Information, 41:2, 198-208, DOI: 10.1080/09613218.2013.769145

To link to this article: http://dx.doi.org/10.1080/09613218.2013.769145

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
 BUILDING RESEARCH & INFORMATION , 2013
Vol. 41, No. 2, 198– 208, http://dx.doi.org/10.1080/09613218.2013.769145

RESEARCH PAPER

Construction costs comparison between

‘green’and conventional o⁄ce buildings
Michael Rehm1 and Rochelle Ade2

1
Department of Property, School of Business and Economics, University of Auckland,12 Grafton Road,
Auckland 1142, New Zealand
E-mail: [email protected]
Downloaded by [University of Toronto Libraries] at 14:44 10 August 2014

2
Ade Consultants,119 First View Avenue, Beachlands, Auckland 2018, New Zealand
E-mail: [email protected]

The actual construction cost of certified green office buildings in New Zealand is compared with conventional buildings.
Although a large body of research exists on the financial and environmental benefits of green buildings, there is little
evidence on capital cost implications for building green. This study serves as the first empirical study to analyse
detailed cost plan data in New Zealand to quantify the impact of green building on construction cost. Data from 17
Green Star NZ v1-certified office buildings were paired with a set of modelled cost estimates derived from the Davis
Langdon Blue Book and the Rawlinsons New Zealand Construction Handbook (the authoritative published sources
for New Zealand construction cost data). The paired data were analysed across five panels using the non-parametric
Wilcoxon matched-pairs signed ranks test. When benchmarked against the modelled costs, green building
construction costs were higher on average, but the difference was not statistically significant. This was true across all
five panels tested: the entire green building dataset, mid-rise buildings, high-rise buildings, 4 Green Star-rated
buildings, and 5 and 6 Green Star-rated buildings. Each panel featured buildings that were above comparative costs
as well as several whose actual costs were below modelled estimates.

Keywords: capital cost, construction, cost, green building, Green Star, office buildings, sustainability

Le coût de construction réel des immeubles de bureaux certifiés verts en Nouvelle-Zélande est comparé à celui des
immeubles classiques. Bien qu’il existe un important corpus de recherche sur les avantages financiers et
environnementaux des bâtiments verts, il y a peu d’éléments probants sur les implications en termes de coût du
capital pour la construction de bâtiments verts. Cette étude constitue la première étude empirique qui analyse les
données des plans détaillés des coûts en Nouvelle-Zélande pour quantifier l’impact du bâtiment vert sur les coûts de
construction. Les données provenant de 17 immeubles de bureaux certifiés Green Star NZ v1 ont été couplées à une
série d’estimations modélisées des coûts tirée du Livre Bleu de David Langdon et du Rawlinsons Manuel de
Construction de Nouvelle-Zélande (les sources publiées qui font autorité concernant les données relatives aux coûts
de construction pour la Nouvelle-Zélande). Les données couplées ont été analysées sur cinq panels en utilisant le test
non paramétrique des rangs signés de Wilcoxon pour échantillons appariés. Lorsqu’ils ont été comparés aux coûts
modélisés, les coûts de construction des bâtiments verts étaient supérieurs en moyenne, mais la différence n’était pas
statistiquement significative. Ceci s’est vérifié sur la totalité des cinq panels testés : l’ensemble des données des
immeubles verts, les immeubles de hauteur moyenne, les immeubles de grande hauteur, les immeubles classés Green
Star 4, et les immeubles classés Green Star 5 et 6. Chaque panel comprenait des immeubles qui se situaient au-dessus
des coûts comparatifs, ainsi que plusieurs dont les coûts réels étaient inférieurs aux estimations modélisées.

Mots clés: coût du capital, construction, coût, immeuble vert, Green Star [Etoile Verte], immeubles de bureaux,
durabilité

# 2013 Taylor & Francis

 Construction costs comparison

Introduction incremental costs associated with the process of achiev-

Buildings are substantial CO2 emitters and contribute ing a green building rating. This involves both appli-
to climate change. A recent study carried out by the cation costs as well as additional consulting required
United Nations Environment Programme (UNEP) under the various rating tools (Cupido, Baetz, Pujari,
suggested that the residential and commercial building & Chidiac, 2010; Owen, 2006). Other studies such
sectors are responsible for approximately one-third of as Malin (2000) have considered the component costs
the greenhouse gas emissions of developed countries of green buildings. Malin investigated cost premiums
(UNEP, 2009), and the property sector’s impact on associated with particular green materials and found
the environment is not limited to carbon emissions that although they have higher initial costs, the added
and energy, with buildings in the European Union con- expense is not severe if the life cycle cost (LCC) is con-
suming 16% of potable water, 50% of raw materials sidered. This accounting technique, which considers
and accounting for 40% of solid landfill waste green building costs and benefits over a building’s
(Keeping, Dixon, & Roberts, 2007). Green building entire life, is discussed by Cole & Sterner (2000,
is viewed as a method of responding to these environ- p. 369) who stated that although LCC accounting is
mental issues. To facilitate this, many organizations, superior to initial capital cost alone it:
such as the New Zealand Green Building Council,
Downloaded by [University of Toronto Libraries] at 14:44 10 August 2014

have introduced rating tools to recognize and remains a limited approach to account for the
improve the sustainability of the building stock. broader environmental and social costs associ-
ated with buildings.
The era of rating tools commenced in 1990 with the
introduction of the Building Research Establishment Cole and Sterner reasoned that these broader cost
Environmental Assessment Method (BREEAM) in the savings associated with the construction of green build-
UK. This was followed in 2000 by the US tool: Leader- ings should be factored in along with operating benefits
ship in Energy and Environmental Design (LEED). The such as lower energy and water consumption.
Green Building Council of Australia was founded in
2002 and used BREEAM and LEED as the basis for Some of these ‘social costs’ such as greenhouse gas
its Green Star rating system, which is specific to the emissions have since been monetized through carbon
Australian environment and building practices pricing, although these externalities may still be
(Madew, 2006). More recently the New Zealand priced too low. More recent LCC research by Kneifel
Green Building Council was established in 2006 with (2010) suggests that by implementing energy efficiency
its first rating tool, Green Star NZ v1, launched in measures commercial buildings can reduce their
2007 (for a detailed history of international rating carbon footprint by 16% on average, thereby improv-
tools, see Reed, Bilos, Wilkinson, & Schulte, 2009). ing green buildings’ life cycle cost-effectiveness.
Despite the overwhelming evidence that buildings
negatively impact the environment, the adoption of Some argue that if integrative design techniques are
green building practices still faces headwinds. The properly employed, green buildings should cost less
primary obstacle is the perception that is costs more than conventional counterparts. In their research
to build green. paper, Hydes & Creech (2000) used two case studies
to demonstrate the potential to achieve green objec-
tives such as increasing the thermal and lighting effi-
Background ciency of the building in ways that reduce, rather
The bulk of the literature on green buildings focuses on than increase, initial capital costs. The authors also
the benefits to building owners, society at large, the implored architects and engineers to be innovative
environment and, in particular, building users (for a and daring by considering the use of salvaged building
recent survey of user perception and satisfaction, see materials and equipment the further to lower both
Baird, 2010). However, a much smaller portfolio of embodied energy and initial cost.
research has investigated the obstacles to building
green such as higher construction costs, with the Other researchers have claimed that higher green
work of Kats, Alevantis, Berman, Mills, & Perlman building costs are mere illusions. Bartlett & Howard
(2003) and Kats et al. (2010) being the most cited. (2000, p. 324) asserted that quantity surveyors:
According to Hwang & Tan (2012) perceived cost pre-
miums remain the main hindrance to the uptake of seriously overestimate the capital costs of energy
green buildings. Of the research on cost implications efficient measures and seriously underestimate
of sustainable construction, most studies explore the potential for cost savings
theoretical issues surrounding green building costs
rather than explicitly testing for cost premiums. thereby misleading decision-makers. Bordass (2000)
shared this sentiment and argued that market partici-
Some researchers have argued that green building soft pants’ perceptions of higher construction costs are
costs are higher than conventional projects due to unsubstantiated. However, with the exception of

199
 Rehm and Ade

Table 1 Summary of perception survey ¢ndings on green building cost savings/premiums

Author(s) (year) N Cost savings (%) None (%) Cost premiums (%)

1^ 5% 6^10% Over 10%

Park, Nagarajan, & Lockwood (2008) 16 0 0 37 38 25

Breslau (2007) 414 1 8 38 30 22
Building Design+Construction (2007) 631 1 13 24 27 35
Ahn & Pearce (2007) 87 1 15 23 34 27

owner-occupied non-residential buildings, commercial Nevertheless, 90% of respondents indicated that

buildings tend to be multi-tenanted and feature leases there is a green building cost premium.
that are ‘net’ of operating expenses. Therefore, the
Downloaded by [University of Toronto Libraries] at 14:44 10 August 2014

revenue benefits from a green building’s energy Building Design + Construction (2007) found that of
savings flows to the tenants rather than the building 631 survey respondents, 86% believed green buildings
owner who has funded the efficiency measures. cost more, 13% stated that they cost the same and 1%
felt that green buildings are less expensive. The mode
In terms of research that has explicitly investigated response was that there exists a green cost premium
green building costs, four surveys have been conducted of more than 10%. Ahn & Pearce (2007) recorded
to date that studied perceived cost premiums. These very similar results from a survey of 87 construction-
surveys asked respondents the degree to which they related firms. In summary, although the estimated
believed green building costs differed from convention- green building cost premium varies across the different
al buildings. Table 1 summarizes these surveys’ perception studies, these studies clearly indicate that
findings. US property professionals hold the view that green
buildings cost more than non-green buildings.
The Deloitte study by Park, Nagarajan, & Lockwood
(2008) reported that none of the survey’s 16 respon- In addition to perception surveys, several studies have
dents felt that green buildings cost the same or less analysed actual or modelled green building construc-
than non-green buildings. All respondents perceived tion cost data in order to determine prevailing cost
green buildings as costing more, with the mode differences. As with the perception studies, this
response being that there exists a cost premium of research has primarily been carried out by professional
between 6% and 10%. Breslau (2007) presented the services firms rather than academic researchers. Conse-
results of a similar perception survey conducted by quentially, very few of these studies are transparent in
Jones Lang LaSalle and CoreNet Global. Unlike the terms of methodology, few subject their data to mean-
Deloitte study, 8% of survey respondents felt that the ingful statistical tests, and none has been subjected to a
cost premium was neutral, while 1% perceived that rigorous peer-review process. Rather, the analytical
green buildings cost less to design and build. research to date has either come from government-

Table 2 Summary of UK BREEAM cost study

Author (year) Cost premium estimation Number of Building type Green building cost premiums
method green (%)
buildings
Good Very Excellent
good

Rawlinson (2007) Actual green building costs 1 O⁄ce building Not studied 2.8
with green-speci¢c
costs itemized
Building Research Establishment (BRE), Actual green building costs 4 O⁄ce building, 0^0.22 0.1^ 5.7 3.3
Centre for Sustainable Construction & against model costs to house and
Cyril Sweett Sustainability and Cost Building Regulations healthcare
ConsultingTeams (2005) standards

200
 Construction costs comparison

commissioned reports or was undertaken by pro- equipment, etc.) with soft costs (design, project man-
fessional services firms actively involved in green build- agement, compliance, government levies, etc.) typically
ing projects. The most noteworthy of these firms is omitted.
Davis Langdon, a global quantity surveying firm, that
has published much of the analytical research into In terms of the comparison methods used in the above
green building costs. studies the predominant approach has been to
compare modelled green building costs against mod-
Tables 2 –4 provide a summary of the quantitative elled conventional building costs. A minority of
research conducted in the United Kingdom, United studies has analysed actual cost data with the
States, Australia and New Zealand that has investi- studies by Matthiessen & Morris (2004, 2007) and
gated the impact of building green on construction Davis Langdon (2009) being the only research that
costs. As with the present study, the focus is on hard has compared actual green building costs with
construction costs (labour, materials, installed actual non-green building data. All three studies

Table 3 Summary of US LEED cost studies

Downloaded by [University of Toronto Libraries] at 14:44 10 August 2014

Author(s) (year) Cost premium estimation Number of Building type Green building cost premiums (%)
method green
buildings
Certi¢ed Silver Gold Platinum

Kats et al. (2010) Survey respondents’ 37 O⁄ce 1.20 2.25 3.37 7.66
(primarily architects) buildings
stated ‘green premium’ 29 Schools 0.35 1.00 1.30 9.60
estimates 17 Academic 1.65 1.80 1.93 2.53
buildings
Davis Langdon Unpaired t-test of actual 12 O⁄ce interior No statistically signi¢cant cost di¡erence
(2009) green ¢t-out costs against ¢t-outs
13 non-green ¢t-out costs
Matthiessen & Morris Unpaired t-test of actual 17 Academic No statistically signi¢cant cost di¡erence
(2007) green building costs buildings
against non-green building 26 Laboratories
costs within each building 25 Libraries
type
Unpaired t-test of actual 15 High-rise
green building costs apartments
against 22 non-green
building costs
Nilson (2005) Actual green building costs 1 O⁄ce building Not studied 0.82 Not studied
with green-speci¢c costs
itemized
Steven Winter Model green building costs 2 O⁄ce building 1.4 ^ 2.1 3.1^ 4.2 7.8^8.2 Not studied
Associates (2004) against model costs to and
‘federal design courthouse
requirements’
Matthiessen & Morris Unpaired t-test of actual 45 Academic No statistically signi¢cant cost di¡erence
(2004) green building costs buildings,
against 98 non-green laboratories
building costs across all and
building types libraries
Kats, Alevantis, Survey respondents’ 33 Schools and 0.66 2.11 1.82 6.50
Berman, Mills, & (primarily architects) o⁄ce
Perlman (2003) stated ‘green premium’ buildings
estimates
Northbridge Meta-analysis of ‘secondary 1 Generic, Soft costs: 1.5^ 3.1
Environmental research’ and unspeci¢ed ¢ctitious Hard costs: 3^8
Management analysis of actual cost data building
Consultants
(2003)
Packard Foundation Model green building costs 1 Proposed, 0.9 13.1 15.5 21.0
(2002) against model ‘market’ cost new o⁄ce
building

201
 Rehm and Ade

Table 4 Summary of Australasian Green Star cost studies

Author(s) Comparison method Number of Building type Green building cost premiums (%)
(year) green
buildings
4 Star 5 Star 6 Star

Fullbrook & Actual ecological sustainable 1 Generic, ¢ctitious 1.25 (4 Star 4.37 (5 Star 6.23 (5 Star
Woods development (ESD) costs government building) building) building)
(2009) against non-green model o⁄ce ¢t-out 2.91
costs (unrated
building)
Davis Langdon Model green building costs 1 Generic, ¢ctitious 0 3^ 5 9^11+
(2007) against generic, non-green o⁄ce building
model costs
Fullbrook (2007) Model green building costs from 20 O⁄ce buildings 3 (Class A) 7 (Class A) Not studied
developer proposals against 7 (Class B) 15 (Class
non-green model costs B)
Downloaded by [University of Toronto Libraries] at 14:44 10 August 2014

Fullbrook, Actual ‘ecologically sustainable 1 Academic building ^15 (savings)

Jackson, & development’ costs against 1 Hospital 1.5
Finlay (2005) non-green model costs 1 School 5.7
1 Library 4.9
Model ESD o⁄ce building costs 1 O⁄ce building Low/medium ESD: 6.5
against non-green model Medium/high ESD: 11.5
costs

were produced by Davis Langdon employees and all reflects the opinions of the survey recipients who
concluded that there is no statistical evidence that elected to respond.
green buildings cost more than non-green buildings.
All other studies have found that green buildings These respondents were predominantly the principal
tend to cost more and the premiums increase with architects of the LEED-certified green buildings tar-
higher rating levels. The highest recorded premiums geted by the survey and are not likely to be impartial
were found in the earliest analytical study by the in their views towards building green. It would be in
Packard Foundation (2002) with LEED Platinum- the interest of such respondents, who have experience
rated buildings incurring a 21% premium. However, in designing green buildings and a strong desire to be
most found lower premiums within 10% of conven- instructed on further LEED-certified projects, to err
tional building costs. on the lower end of the green cost premium spectrum.
This is particularly the case since all respondents are
The most cited studies on green building cost pre- named in footnotes throughout Kats et al. (2003)
miums are Kats et al. (2003, 2010). As with many and within Appendix B of Kats et al. (2010). It is con-
other studies, Kats et al. analysed several building ceivable that a respondent associated with projects
types including offices but they employed an entirely featuring low green building cost premiums would
different approach to measuring cost premiums. be more likely to be instructed by a would-be green
Rather than comparing actual or modelled costs, building developer than would a respondent linked
Kats et al. conducted a mail questionnaire survey to projects with high cost premiums. Neither of Kats
with a single question responsible for quantifying et al.’s publications clarify whether or not respondents
cost premiums, which asked: were notified that their names would be disclosed in
the subsequent research publication, but given that
What was the additional cost incurred (the cost this is the case it may be reasonable to assume that
premium) to construct a green building vs. a con- one of the possible incentives for participating was
ventional building? promotion of their services in the publication. In
addition to issues of respondent bias, the authors’
This method is problematic for a number of reasons. choice of directing the surveys to architects rather
Firstly, the survey itself would be subject to self-selec- than quantity surveyors to ascertain green building
tion bias. In the case of the 2010 study, over 300 cost premiums is also somewhat problematic as cost
surveys were mailed but only 170 responses were estimation is generally not a core competency of prac-
received, seven of which did not reply to the cost tising architects whereas quantity surveyors specialize
premium question. Therefore, the analysis on cost in this very task.

202
 Construction costs comparison

Kats et al. (2010) included in their Appendix C the owner, a general contractor and a project manager.
entire green building dataset. The present analysis of Each of the interviewees had first-hand experience
the Kats et al. data found that 24% of Certified, with green buildings. The aim of the interviews was
26% of Silver and Gold, and 18% of Platinum-rated to understand the current perceptions of green building
buildings were recorded as incurring no green construction costs and to gauge whether the introduc-
premium. The cost premiums from Kats et al. pre- tion of Green Star had affected New Zealand office
sented in Table 3 are derived from their Appendix C markets. The insights gained from the interviews
and represents mean premiums across the three main were drawn upon by the authors when interpreting
property types investigated: office buildings, schools the empirical results.
and academic buildings.
Table 5 indicates the interviewees’ perceived green
The present study’s research design addressed the issue building cost premiums. Given the choice of between
of bias by electing to acquire and analyse actual cost five potential responses ranging from cost savings to
data from quantity surveyors’ proprietary cost plans a cost premium over 10%, none of the interviewees
for each of the green buildings studied. As a condition responded that a Green Star-rated project would cost
of disclosing such proprietary information, the authors less than or equal to a conventional building. Most
Downloaded by [University of Toronto Libraries] at 14:44 10 August 2014

agreed with each data provider that individual build- felt that 4 Green Star buildings would incur cost pre-
ings would not be identified. It was largely this level miums of up to 5% while none estimated that such a
of ensured confidentiality that enabled access to sensi- project would be subject to a cost premium of over
tive commercial cost plans. 10%. Similar to past perception studies, the cost
premium varied by rating level with 5 Green Star build-
The studies that have investigated green building costs ings largely perceived to attract cost premiums of
in Australasia include three studies commissioned by 6– 10% while 6 Green Star projects would tend to
the New Zealand government and a research report attract premiums in excess of 10%.
by Davis Langdon focused on Australian office
markets. Of these the only study to analyse actual Interviewees were also asked why green buildings cost
cost data was Fullbrook, Jackson, & Finlay (2005) more to construct. The reasons indicated included the
which predates New Zealand Green Star and investi- fact that sustainable features such as green materials,
gated four recently completed ‘ecologically sustainable high-performance cladding systems, rainwater har-
developments’. Their research found that the build- vesting and energy-efficient mechanical equipment
ings’ actual hard costs ranged from being 15% below could increase capital costs. They also noted the soft
non-green modelled costs to roughly 6% more expens- costs associated with applying for and satisfying the
ive. The technique used to quantify cost premiums used requirements of Green Star. Several interviewees
in Fullbrook et al.’s study is similar to that by the stated that since Green Star entered the market, build-
Building Research Establishment (BRE), Centre for ing quality in general had risen. More specifically
Sustainable Construction & Cyril Sweett Sustainability these interviewees felt that Green Star-rated buildings
and Cost Consulting Teams (2005) which also took a tended to built to higher specifications of which many
case study approach and analysed the actual costs of did not necessarily benefit the environment. Such
four BREEAM-certified buildings against modelled high-quality design features were in addition to
costs of the same type of buildings constructed to green building materials and systems put in place to
Building Regulations standards. The methods of com- earn a particular Green Star rating. Furthermore,
parison between green and non-green construction the interviewees felt that this enhanced level of
costs used in these two 2005 studies are the most quality was particularly evident in the case of urban
similar to the present study. The main difference is fringe developments which tended to consist of low-
that the present paper investigates 17 commercial
office buildings whereas Fullbrook et al. and the BRE
et al. analysed a handful of individual case studies,
Table 5 Perceived New Zealand green building cost savings/
each with a different use and nearly all case studies
premiums
of public, non-commercial facilities.
Rating Cost None Cost premiums (%)
savings (%)
(%)
Methods 1^ 5% 6^10% 10%1
The research was conducted in two stages: qualitative 4 Green Star 0 0 75 25 0
interviews followed by quantitative data analysis.
Initially a series of semi-structured interviews was 5 Green Star 0 0 31 62 8
held with 15 industry professionals from across New 6 Green Star 0 0 0 40 60
Zealand. This included six engineers, three quantity
surveyors, two architects, a developer, a building Note: N¼15 industry professionals.

203
 Rehm and Ade

and mid-rise buildings and up until the introduction comprehensive construction cost index. Within the
of Green Star rating tools had historically been of Introduction the editor asserts that:
lower quality than high-rise office buildings in
central business districts. the Handbook has achieved a reputation as the
leading authority on the various aspects of con-
struction costs in New Zealand, and is recog-
Quantitative research nised as an authoritative text in the courts for
cases relating to disputes in respect of construc-
The second stage of the study involved the collection
tion cost.
and analysis of actual green building cost data. With
(Giddens, 2010)
a focus on actual costs it was decided to target New
Zealand office buildings that had been certified under
While the Rawlinsons Handbook is a commercial
the New Zealand Green Building Council’s inaugural
publication, current and past Blue Books are available
rating tool: Green Star NZ v1. Throughout the
for free download from the Davis Langdon website
country there are 49 Green Star NZ v1-certified pro-
(http://www.davislangdon.com/ANZ/Research/). As
jects. The developer and/or owner of each building
previously mentioned, the authors’ adopted approach
Downloaded by [University of Toronto Libraries] at 14:44 10 August 2014

was solicited for participation in this study. Twenty-

of comparing actual costs to model costs aligns the
one owners, roughly 43%, agreed to contribute to
present study with the comparison methods employed
the study and their respective quantity surveyors
by Fullbrook et al. (2005) and BRE et al. (2005).
involved with each development provided access to
the full project cost plan.
Both the Blue Book and Rawlinsons Handbook are
silent in regard to the specific methodology employed
Four of these 21 buildings were refurbishments rather
to determine their cost estimates. Both are also silent
than new construction and were therefore excluded,
in terms of whether or not green building construction
resulting in an overall dataset of 17 buildings. Of the
costs are incorporated. To clarify the methodologies
cost plans analysed, two were estimated costs as the
used, the authors contacted the editor of Rawlinsons
projects were under construction when the data were
Handbook who confirmed that its figures are based on
collected. However, the remaining 15 cost plans pro-
modelled costs rather than a sampling of actual whole
vided final, as-built costs. Table 6 provides a break-
building construction costs and that they reflect prevail-
down of these projects by height category and Green
ing costs for conventional buildings rather than green
Star rating. The dataset includes eight 4 Green Star
buildings. In contrast, the cost estimates published in
buildings, eight 5 Green Star and one 6 Green Star
the Blue Book are derived from tender prices rather
building. The majority of buildings were mid-rise
than Rawlinsons Handbook’s modelled costs that
(under six stories). In terms of location, 12 of the build-
track changes in construction labour and material prices.
ings were located in Auckland with the balance found
elsewhere in the country. The green buildings studied
In terms of the incorporation of green building costs the
ranged in initial year of construction from 2006 to
2010 Blue Book (Davis Langdon, 2010) stipulates that:
2010.
all rates generally relate to a building achieving a
The authors initially sought to gather actual cost data
4 green star rating and a 4.5 NABERS energy
on non-green office buildings for use as a baseline for
base building rating.
comparison as per Matthiessen & Morris (2004,
2007). However, this proved unfeasible. Therefore,
This reference is derived from the Property Council
the research was redesigned to compare actual costs
of Australia’s (2006) A Guide to Office Building
with modelled costs derived from Davis Langdon’s
Quality. It states that new Australian office buildings
Blue Book and the Rawlinsons New Zealand Con-
will only be classified as ‘Premium’ or ‘Class A’ if they
struction Handbook, which is the country’s only
achieve these stated green ratings or higher. The
updated 2012 guide (Property Council of Australia,
Table 6 Frequency table of analysed green buildings by height 2012) is even more specific with ‘Premium’ class build-
and rating ings requiring either a 5 Green Star or 5 star NABERS
(National Australian Built Environment Rating
Building height 4 Green 5 and 6 Total System) rating. This means that in the Australian
Star Green Stars office market, high standard buildings must be green,
which in turns suggests that green buildings are not
Under six stories 6 5 11 only sustainable, but also constructed to a high standard
Six stories and 2 4 6 in general. Based on the specifications of the analysed
over green buildings evident in the cost plans, the authors
Total 8 9 17 believe that this green – high standard nexus exists in
New Zealand office markets as well.

204
 Construction costs comparison

In an attempt to gauge the accuracy of cost information One challenge in comparing actual costs from non-
contained within the Blue Book and Rawlinsons uniform cost plans with modelled costs from the Blue
Handbook, cost data were compared between the Book is that the scope of some green buildings differ.
two sources. Specifically, the Blue Book’s ‘medium- For instance some developments include integrated
rise standard offices’ costs were paired with Rawlin- fit-outs of tenanted spaces while others are constructed
sons Handbook’s ‘3 to 5 storey office buildings with as ‘shell-and-core’ with tenants being responsible for
A/C’ and ‘high standard [high-rise] offices’ costs their own fit-out. Since the cost plans were largely
were compared against figures for ‘6 to 15 storey non-uniform as they were supplied by various quantity
office buildings with A/C’. The variation in the two surveying firms, some cost plans made it difficult to
publications’ mean office building construction costs isolate interior fit-out costs associated with lease
for Auckland and Wellington from 2006 to 2009 is spaces versus common areas. However, costs for
roughly 5%. However, the publications’ high-rise internal finishing in general were clearly isolated in
office figures diverge in 2010 with Rawlinsons Hand- each of the cost plans.
book including slight price increases from the previous
year while that year’s Blue Book records a considerable Since the Blue Book is limited in its provision of cost
drop in high standard office building hard costs putting breakdowns by building elements, the Rawlinsons
Downloaded by [University of Toronto Libraries] at 14:44 10 August 2014

it on par with its 2006 figures. This was unlikely to be a Handbook’s section on ‘Elemental Costs of Buildings’
typographical error on the part of Davis Langdon, was used to quantify the internal finishing component
given that the 2011 Blue Book further reduced high of the modelled costs. This was done by multiplying
standard office costs. Instead, the drop in construction the Blue Book per square metre costs by the respective
prices likely mirrors a drop in tender prices, which Rawlinsons Handbook’s percentage share of costs
Rawlinsons Handbook’s methodology fails to capture. associated with internal finishing. This amount was
then removed to arrive at hard costs for ‘shell-and-
Importantly, only the Blue Book explicitly provides an core’ office buildings less the costs associated with
indication of cost differences between high standard internal finishes. Lastly several other actual costs
and average standard office buildings. As discussed were removed from cost plans to align them with the
above, in Australasia green buildings tend to be high- Blue Book’s costs which are defined as:
standard buildings and therefore high-standard con-
struction. This is specifically the case for the popu- inclusive of builders preliminaries and profit but
lation of 17 green buildings being studied. Therefore, exclusive of site works, external services, land
cost estimates for high-standard buildings are needed and interest costs.
to establish whether or not green buildings are
subject to cost premiums. Given the Blue Book’s recog- With the necessary adjustments made to actual and
nition of differences in standard and the fact that its modelled costs, the data were tested statistically to
methodology appears to reflect better movements in determine if the green buildings’ costs were indeed sig-
tender prices, the authors have chosen it as the main nificantly higher than the modelled costs. Since the
source for producing model construction costs. number of observations per tested group ranged
between six for the set of high-rise buildings and 17
Since the Blue Book provides high-standard costs for for the overall dataset, the use of a parametric test
only high-rise buildings (over ten stories), an implied such as the paired Student’s t-test was not feasible
high standard ‘premium’ can be derived by computing since such sample sizes were below the commonly
the cost difference between ‘High Standard [High-rise] accepted threshold of 30 observations (Cohen, 1988;
Office’ and ‘Average Standard High-rise Office’. This Wilson van Voorhis & Morgan, 2007). That said,
amount can then be added to average standard low- the t-tests conducted by Matthiessen & Morris
and mid-rise costs to approximate the high-standard (2007) and Davis Langdon (2009) violated this
costs for these categories of buildings, which account threshold. It was therefore decided to use the non-para-
for 15 of the 17 green buildings analysed. metric Wilcoxon matched-pairs signed ranks test as an
alternative to the t-test (for an explanation of the test
Once the model cost source was determined the next procedure, see Siegel, 1956, pp. 75 – 83). The test con-
step was to pair individual green buildings with their siders a null hypothesis of no cost difference between
respective annual Blue Book model costs. For consist- the actual green building cost and Blue Book model
ency, the year of initial construction was used to deter- cost, with the alternative hypothesis being that a
mine the handbook from which modelled costs would green building cost premium exists.
be derived. Aside from building scope (e.g. three to ten
storey mid-rise) cost estimates are given for three New From the semi-structured interviews, it was suggested
Zealand cities: Auckland, Wellington and Christch- that aside from their provision of sustainable materials
urch (from 2010 onwards). In the seldom case when and systems, Green Star-rated buildings tend to be of
an analysed green building project is located in a higher overall quality in comparison with non-green
different city, the costs for the nearest city were used. buildings. Therefore, the hypothesis testing compares

205
 Rehm and Ade

Table 7 Summary results of Wilcoxon matched-pairs signed ranks tests

Panel Description N Above model cost Below model cost W z pa

I All buildings 17 10 7 42 1.63 0.102

II Three to ¢ve stories 11 7 4 15 1.60 0.110
III Six to 15 stories 6 3 3 8 0.52 0.600
IV 4 Green Star 8 4 4 14 0.56 0.575
V 5 and 6 Green Stars 9 6 3 10 1.48 0.139

Note: aDirectional, one-tailed signi¢cance level ( p-value).

actual costs with the Blue Book model cost assuming a Green Star buildings. Unfortunately, the present
high standard specification. dataset of only 17 observations does not allow for this.
Downloaded by [University of Toronto Libraries] at 14:44 10 August 2014

Table 9 presents summary statistics of the difference

Results between actual and model costs. The ranges and stan-
The summary results of the Wilcoxon matched-pairs dard deviations are substantial and reflect the wide
signed ranks tests are given in Table 7. It provides variation typical of commercial construction costs.
the number of green buildings above and below the Matthiessen & Morris (2007) noted that such cost var-
Blue Book model cost. It also indicates the Wilcoxon iance reflects the heterogeneity of the commercial
value (W), z-score and p-value for each test. None of building stock. It is for this reason that the present
the panels studied returned statistically significant authors elected to analyse their green building dataset
differences between actual and modelled construction by height category and sustainability rating. Such an
costs. All the panels featured a mix of observations approach was not taken by Matthiessen and Morris
above and below model cost, with six to fifteen who instead elected to aggregate all green buildings
storey and 4 Green Star panels having an equal into a single population regardless of individual
number of buildings on either side of the benchmark. rating levels. Moreover, unlike the present study, Mat-
thiessen and Morris made no attempt to subdivide their
Table 8 provides a frequency of cost differences datasets into more homogenous subpopulations based
between actual green buildings and modelled costs on the physical characteristics of the academic build-
across the five panels. It indicates a wide spectrum of ings, laboratories, libraries and apartment buildings
differences. However, the mode range in four of five analysed. In the present study, strictly office buildings
panels is actual costs 0– 9% below model costs. The are considered.
exception is 5 and 6 Green Star buildings where the
largest cohort features actual costs more than 50% Despite these efforts the construction costs vary greatly
above model costs. The variation in costs within within cohorts with the largest difference occurring
panels indicates a considerable degree of heterogeneity within the panel of 5 and 6 Green Star buildings. In
among green buildings. Some of this could potentially this case one building’s actual hard cost was 35%
be addressed through a larger dataset with more pre- below the Blue Book modelled costs while another
cisely defined panels such as three to five storey, 4 building in the same group was roughly 96% higher

Table 8 Frequency table of green building cost di¡erences

Panel Description N Actual cost below model Actual cost above model cost
cost

20%1 10^19% 0^9% 0^9% 10^19% 20^ 29% 30^39% 40^ 49% 50%1

I All buildings 17 2 0 5 2 1 1 1 2 3
II Three to ¢ve stories 11 1 0 3 1 1 0 1 2 2
III Six to 15 stories 6 1 0 2 1 0 1 0 0 1
IV 4 Green Star 8 1 0 3 1 1 0 0 2 0
V 5 and 6 Green Stars 9 1 0 2 1 0 1 1 0 3

206
 Construction costs comparison

Table 9 Summary statistics of green building cost di¡erences

Panel Description N %

Mean Median Standard Deviation Minimum Maximum

I All buildings 17 18.2 7.2 35.2 ^ 35.4 96.4

II Three to ¢ve stories 11 23.3 14.5 36.4 ^ 24.3 96.4
III Six to 15 stories 6 8.7 2.3 33.9 ^ 35.4 65.3
IV 4 Green Star 8 8.1 0.8 23.9 ^ 24.3 45.2
V 5 and 6 Green Stars 9 27.2 24.4 42.2 ^ 35.4 96.4

than estimated costs. Even the tightest cost range is Kingdom. Collecting actual cost data from such
Downloaded by [University of Toronto Libraries] at 14:44 10 August 2014

considerably large. Based on standard deviation, sizable markets should result in a considerably larger
Panel IV, consisting of 4 Green Star buildings, is the green building dataset. With more observations gath-
most closely clustered about the mean difference ered, more homogenous panels can be created. This
between actual and model cost. Aside from the wide was a limiting factor in the present study which had
variation of individual high-rise costs, the overall only 17 total observations. The factors by which
Panel III and IV differences between actual and panels can be defined could include combinations of
model costs are exceptionally small with a mean differ- rating level and building scope (e.g. 4 Green Star high-
ence around 8% and median cost differences of 2% rise buildings) as well as building quality based on
and 1%, respectively. This along with the Wilcoxon expert classifications (e.g. Property Council of Austra-
test results suggests that, overall, green buildings in lia’s office building classes) or market rental rates.
New Zealand are not systematically more costly to Such finer panel definitions would help reduce cost vari-
construct than conventional buildings. ation within each analysed group. Ideally each panel
would include at least 30 observations to enable the
Conclusions use of more robust, parametric statistical tests.
This study provides the first in-depth investigation into
actual green building construction costs in New With more observations future research can also attempt
Zealand based on detailed cost plans. With initial to determine whether or not particular delivery methods
capital costs being the primary obstacle to the uptake influence green building costs. For instance, several
of green buildings, research of this nature is vital researchers have argued that by embracing integrative
towards understanding whether or not green buildings design, green buildings should cost less than conventional
cost more to construct. Based on the comparison of 17 buildings. The use of such holistic approaches is referred
green buildings’ actual cost data against modelled cost to as ‘tunnelling through the cost barrier’ by Hawkins,
estimates, it is concluded that, on the whole, green Lovins, & Lovins (2010, pp. 113–124) who asserted
buildings are not inherently more expensive due to that an integrative design process can produce enough
their provision of sustainable materials and systems. energy savings that a building’s costly mechanical
Although the green buildings analysed have shown system can be downsized or even eliminated, thereby
higher costs on average, a sizable portion of buildings more than offsetting the additional costs associated
have been found to be below modelled costs. As noted with the efficiency measures (e.g. double-glazing). Such
by Matthiessen & Morris (2007) such cost variance claims could be verified if a sufficiently large dataset
reflects the heterogeneity of the commercial building were gathered and information on each project’s delivery
stock. Interestingly the largest variance is found in the method was available. Lastly, future research agendas
cohort of 5 and 6 Green Star buildings. This group of can also investigate where cost premiums reside within
nine buildings contains the largest cost savings a cost plan. Knowing which building elements tend to
(– 35%) and highest cost premium (96%). This alone attract premiums, or savings, will allow these costs to
captures the sentiment of Hydes & Creech (2000) and be effectively addressed through cost management tech-
other researchers that high-performing green buildings niques and possibly through use of innovative project
do not necessarily need to cost more. In fact they can delivery methods such as integrative design.
be considerably less expensive, particularly when mech-
anical systems can be minimized or omitted.
Acknowledgements
Future research can address the issue of cost variance by The authors thank the anonymous referees for their
performing similar analyses on larger office markets insightful feedback which benefitted the analysis and
such as Australia, the United States or the United this paper.
207
 Rehm and Ade

References Kats, G., Alevantis, L., Berman, A., Mills, E., & Perlman, J.
(2003). The costs and financial benefits of green buildings:
Ahn, Y., & Pearce, A. (2007). Green construction: contractor
A report to California’s Sustainable Building Task Force.
experiences, expectations, and perceptions. Journal of
Retrieved May 4, 2012 from: http://www.usgbc.org/Docs/
Green Building, 2(3), 1 –17.
News/News477.pdf/.
Baird, G. (2010, May). What the users think of sustainable
Kats, G., Braman, J., & James, M. (2010). Greening our built
buildings: a global overview, Paper presented at the Sustain-
environment: Costs, benefits, and strategies. Washington,
able Building Conference, Wellington, New Zealand, 2010.
DC: Island Press.
Bartlett, E., & Howard, N. (2000). Informing the decision
Keeping, M., Dixon, T., & Roberts, C. (2007). The energy per-
makers on the cost and value of green building. Building
formance of buildings directive and commercial property
Research & Information, 28(5/6), 315– 324.
investment, Retrieved May 4, 2012 from: https://www.ipf.
Bordass, B. (2000). Cost and value: fact and fiction. Building
org.uk/home/research/ipf_research_programme_2011-2015/
Research & Information, 28(5/6), 338– 352.
major_projects/ipf_research_reports/energy_performance_
Breslau, B. (2007). Sustainability in the corporate real estate indus-
and_cpi/default.aspx/.
try: Perceptions and trends. Retrieved May 4, 2012 from:
Kneifel, J. (2010). Life-cycle carbon and cost analysis of energy
http://www.joneslanglasalle-boston.com/ma/corporate/rese
efficiency measures in new commercial buildings. Energy
arch/download/BothIssues_Breslau_Sustainability_12-16-
and Buildings, 42(3), 333–340.
07.pdf/.
Madew, R. (2006). Dollars and sense of Green Buildings 2006:
Building Design+Construction. (2007). Green buildings
Building the business case for green commercial buildings
research white paper. Retrieved May 4, 2012 from: http://
Downloaded by [University of Toronto Libraries] at 14:44 10 August 2014

in Australia. Retrieved May 4, 2012 from: http://www.

www.bdcnetwork.com/bdcs-2007-green-buildings-research-
gbca.org.au/uploads/234/1002/Dollars%20and%20Sense%
white-paper-where-building-owners-end-users-and-aec-
20of%20Green%20Buildings%202006.pdf/.
professionals/.
Malin, N. (2000). The cost of green materials. Building Research
Building Research Establishment. (BRE), Centre for Sustainable
& Information, 28(5/6), 408 –412.
Construction & Cyril Sweett Sustainability and Cost Con-
Matthiessen, L., & Morris, P. (2004). Costing green: A comprehen-
sulting Teams. (2005). Putting a price on sustainability.
sive cost database and budgeting methodology. Retrieved May
London: BRE.
4, 2012 from: http://www.davislangdon.com/USA/Research/
Cohen, J. (1988). Statistical power analysis for the behavioral
ResearchFinder/2004-Costing-Green-A-Comprehensive-
sciences (2nd ed.). Hillsdale, NJ: Erlbaum.
Cost-Database-and-Budgeting-Methodology/.
Cole, R., & Sterner, E. (2000). Reconciling theory and practice of
Matthiessen, L., & Morris, P. (2007). Cost of green revisited.
life-cycle costing. Building Research & Information, 28(5/
Retrieved May 4, 2012 from: http://www.davislangdon.
6), 368– 375.
com/USA/Research/ResearchFinder/2007-The-Cost-of-
Cupido, A., Baetz, B., Pujari, A., & Chidiac, S. (2010). Evaluating
Green-Revisited/.
institutional green building policies: a mixed-methods
Nilson, M. (2005). Quantifying the cost impacts of LEED-NC
approach. Journal of Green Building, 5(1), 115–131.
Gold construction in New York City. Easton, PA: (BEng
Davis Langdon. (2007). The cost and benefit of achieving green
(Hons) thesis), Lafayette College.
buildings. Retrieved May 4, 2012 from: http://www.
Northbridge Environmental Management Consultants (2003).
davislangdon.com/upload/StaticFiles/AUSNZ%
Analyzing the cost of obtaining LEED certification. Retrieved
20Publications/Info%20Data/InfoData_Green_Buildings.
May 4, 2012 from: http://www.cleanair-coolplanet.org/for_
pdf/.
communities/LEED_links/AnalyzingtheCostofLEED.pdf/.
Davis Langdon. (2009). The cost of green in NYC. Retrieved
Owen, C. (2006). Reaching for the stars. Architecture Australia,
May 4, 2012 from: http://www.davislangdon.com/upload/
6(95), 113–116.
images/publications/USA/Cost_Study_NYC_10.02.09.pdf/.
Packard Foundation (2002). Building for sustainability report.
Davis Langdon. (2010). The Blue Book 2010. Retrieved May
Los Altos, CA: The David and Lucile Packard Foundation.
4, 2012 from: http://www.davislangdon.com/upload/
Park, C., Nagarajan, S., & Lockwood, C. (2008). The dollars
StaticFiles/AUSNZ%20Publications/The%20Blue%
and sense of green retrofits. Retrieved May 4, 2012 from:
20Book/BlueBook%202010.pdf/.
http://www.deloitte.com/assets/Dcom-UnitedStates/Local%
Davis Langdon. (2011). The blue book 2011. Retrieved May
20Assets/Documents/us_re_Dollars_Sense_Retrofits_190608_
4, 2012 from: http://www.davislangdon.co.nz/upload/
pdf/.
StaticFiles/AUSNZ%20Publications/The%20Blue%20Book/
Property Council of Australia. (2006). A guide to office building
Blue_Book_2011.pdf/.
quality. Sydney: Property Council of Australia.
Fullbrook, D. (2007). Value cases for achieving Green Star NZ 4
Property Council of Australia. (2012). A guide to office building
Star and 5 Star environmental ratings in commercial office
quality (2nd ed.). Sydney: Property Council of Australia.
buildings. Wellington: Ministry for the Environment.
Rawlinson, S. (2007). Sustainability offices. Building Magazine,
Fullbrook, D., & Woods, J. (2009). Value case for Green Star-
Issue 2 (January), 54 –58.
rated fitting out of central government office accommodation
Reed, R., Bilos, A., Wilkinson, S., & Schulte, K. (2009). Inter-
(ME 934). Wellington: Ministry for the Environment.
national comparison of sustainable rating tools. Journal of
Fullbrook, D., Jackson, Q., & Finlay, G. (2005). Value case for
Sustainable Real Estate, 1(1), 1–22.
sustainable building in New Zealand (ME 705). Wellington:
Siegel, S. (1956). Non-parametric statistics for the behavioral
Ministry for the Environment.
sciences. New York, NY: McGraw-Hill.
Giddens, C. (2010). C Giddens (Ed.), Rawlinsons New Zealand
Steven Winter Associates (2004). GSA LEED cost study,
construction handbook. Auckland: Rawlinsons Media.
Retrieved May 4, 2012 from: http://www.wbdg.org/ccb/
Hawkins, P., Lovins, A., & Lovins, L. (2010). Natural capitalism
GSAMAN/gsaleed.pdf/.
(10th ed.). London: Earthscan.
United Nations Environment Programme (UNEP) (2009). Build-
Hwang, B., & Tan, J. (2012). Green building project manage-
ings and climate change: Summary for decision-makers.
ment: obstacles and solutions for sustainable development.
Paris: UNEP.
Sustainable Development, 20(5), 335– 349.
Wilson van Voorhis, C., & Morgan, B. (2007). Understanding
Hydes, K., & Creech, L. (2000). Reducing mechanical equipment
power and rules of thumb for determining sample sizes. Tutor-
cost: the economics of green design. Building Research &
ials in Quantitative Methods for Psychology, 3(2), 43–50.
Information, 28(5/6), 403 –407.

208