The willingness to pay for energy efficient and accessible office buildings

(1)

1

The willingness to pay for energy efficient and accessible office buildings

Abstract

Policy makers are constantly seeking for new ways to reduce the dangers of global warming. As of 2023, commercial properties in the Netherlands with an EU energy label of D-G may no longer be serving as office buildings. This paper examines the pricing effects of two aspects of the comprehensive definition of sustainability: energy efficiency and accessibility. This paper contributes in several ways. Firstly, the little existing Dutch literature that models the impact of office value-drivers uses appraisal or rental data rather than transactional exchange prices. Secondly, this paper is the first to present empirical evidence on investors’ willingness to pay for the level of energy efficiency and alternative accessibility of office buildings in a property market where sustainability has become a growing theme of importance. Thirdly, this thesis is the first to examine whether the pricing difference investors may have been willing to pay for energy efficient office buildings has developed over the past years. For this paper, unique data on 625 office building transactions and characteristics have been gathered from the Randstad conurbation. Hedonic modelling techniques are applied and building, locational, investment, tenancy and market features are controlled to avoid potential omitted variable bias. Findings suggest that, over the period from 2012 to 2017, energy efficient office buildings, on average, have earned a premium of 6.8% relative to similar yet less energy efficient office buildings. Moreover, the premium is found to have increased as time progressed. Investors are also found to pay “green” premiums for the walkability relative to the nearby amenities and the accessibility relative to train stations. In addition to quantitative research, various real estate professionals are asked for their stance on the results of this study, the potential impact that the announced regulation may have had on investor behaviour and more.

Keywords: Energy efficiency, Sustainability, Commercial real estate

JEL classifications: P48, Q01, Q30, Q40, R30, R33

Specialization: MSc Finance (Real Estate Finance track)

Document: Master Thesis

Name: Korse, Felix

Date: July 1st_{, 2018}

Supervised by: dr. F.P.W. Schilder

(2)

2

Statement of originality

This document is written by Felix Korse who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

3

Preface and Acknowledgements

Exactly two years ago, I finished my Bachelor of Science in Economics and Business at the University of Amsterdam. Since I had always been interested in the Real estate sector, I decided to start working for a retail brokerage firm. As my interests and ambitions grew, I pursued my career at Spring Real Estate, working as a real estate market analyst for the capital market department. In September 2017, I started a Master’s in Finance at the University of Amsterdam that allowed me to specialize in real estate finance. After having gained a great amount of knowledge over the past years in over 40 courses, the time has come for the final piece of my educational career at the University of Amsterdam.

I would like to express my gratitude to several people. Firstly, I would like to thank my supervisor dr. F.P.W. Schilder for consistently providing me with quick and constructive feedback regarding my work, which resulted in considerable improvements. Secondly, I would like to thank my colleague Sjors van Iersel, not only for allowing me to use and expand the transactional database of Spring Real Estate, but also for his input and assistance. Thirdly, I am thankful to my managers Alexander Zweering and Robert Kat for providing me with the financial resources that allowed me to retrieve the latest and most reliable data. Furthermore, I would like to thank dr. A.S. Booij for providing me with some econometric insights that improved my estimation model, the Department of Spatial Economics of the VU Amsterdam for their collaboration and the interviewees for their willingness to contribute to my work and provision of interesting insights. Lastly, I am grateful to my parents, Frans and Yvette, brother and sister, Jasper and Veerle, and of course my girlfriend, Robin, for their encouragement, involvement and support.

I hope all of you will enjoy reading this paper as much as I enjoyed writing it. Felix Korse

(4)

4

1. Introduction

This section elaborates on the motivation and relevance of this paper, introduces the research questions and prepares the reader for the main body of this thesis.

1.1. Motivation and relevance

Sustainability has become a growing theme of importance over the past decades. In the Paris Agreement of 2015, 175 countries signed an international climate treaty to decrease the effects of global warming. The potential contribution of the real estate sector to limit global warming seems substantial, as the real estate sector is responsible for 40% of the energy consumption and 36% of the CO2 emissions in Europe (EPRA, 2018). Existing literature acknowledges and confirms the effective possibilities for the real estate sector to reduce global CO2 emissions (Stern, 2008; Eichholtz et al.,

2013).

Over the past 20 years, various initiatives have been taken to stimulate sustainability in the real estate sector. For example, indicating sustainability performance of individual building and asset portfolios by using numerous labelling frameworks (e.g. Energy Star, LEED, BREEAM, EU energy label and GRESB), creating initiative groups (e.g. Green Building Council) and, of course, new policies and regulations.

Given the strong ambitions of CO2 reductions, policy makers are constantly seeking for new regulations to improve environmental performance in the real estate sector. On October 28, 2016, the Dutch Minister of Housing and Civil Service announced his plan to make it compulsory for all Dutch office buildings to be certified with an EU energy label of at least C in 2023 (Ministerie van Binnenlandse Zaken en Koninkrijksrelaties, 2016). The new requirement poses complications for investors who have invested in less environmental friendly buildings with a label D to G, as these objects may no longer be operating as offices. Owners will therefore have to decide whether to make investments to upgrade the level of energy efficiency of the property or not. If owners decide not to invest in refurbishment, the structures will either be demolished or given another use. According to estimates of Arnoldussen et al. (2016), almost 52% or 43 million square meters of the total leasable office floor area in the Netherlands does not meet the 2023 sustainability requirement. They estimated a total required investment of approximately €860 million to make all office buildings meet the new regulations. In reaction to the recently announced policy, one of the largest bank financers of The Netherlands, ING Real Estate Finance, now requires clients who own energy inefficient office properties to submit an appropriate sustainability plan by the end of 2018 (Bak, 2017a).

Although it is intuitive that social benefits arise from sustainability trends and regulations, commercial real estate investors are mainly interested in the financial consequences. It could be that the market value of less energy efficient office buildings are substantially lower due to the required refurbishment costs to meet the future regulations. It is important for investors and owner to know whether the level of energy efficiency is reflected in the capital value of a building for three main reasons. Firstly, a reflection of energy efficiency in building value could incentivize market participants to invest more in energy efficient measures, as this would indicate that investments in refurbishment are more likely to be recovered. Secondly, the findings could help investors understand what additional expense they should be willing to pay when considering a more energy efficient office investment opportunity. Finally, premiums for more sustainable buildings would indicate more ability for these buildings to preserve future value in times of changing sustainability policies and regulations. Outcomes of existing literature on this topic seem consistent: sustainable offices are typically associated with both sales and rental premiums (Devine & Kok, 2015). Most related research has been conducted in the U.S. In contrast, Dutch literature on the financial implications of sustainability is very limited. Although outcomes of existing Dutch literature suggest that office building occupiers in the Netherlands are willing to pay a rental premium for more energy efficient and accessible office buildings (Kok and Jennen, 2012), no academic research has empirically examined yet whether

(6)

6

investors are also willing to pay a premium, let alone since the 2023 sustainability regulation was announced. Moreover, Dutch relevant literature primarily focuses on the residential sector rather than the office sector (Kok and Jennen, 2012). The lack of Dutch literature on the relationship between sustainability and transaction price premiums in today’s office market can be explained by two main reasons. Firstly, existing literature has been hampered by the slow dispersion of heterogeneous sustainability certification schemes (Brounen & Kok, 2010; Kok and Jennen, 2012). As will be described in Section 2.4, the level of sustainability would ideally be measured by either LEED or BREEAM performance certificates, as these frameworks touch upon most aspects that qualify a building as sustainable. However, these schemes are not customary in the Dutch market. Consequently, the most workable certificate to indicate whether an office building qualifies as sustainable or not in the Dutch market is the European Energy Performance Certificate (EPC), which labels a property with an EU energy label. What is important to notice is that, as its name suggests, this framework only focuses on one aspect of the broad sustainability concept: energy efficiency. According to recent estimates1_{, only}

approximately 23% of the Dutch office stock - the equivalent of 18.8 million square meters of leasable floor area - is provided with an EPC, corresponding to solely 6% of the total number of offices (Arnoldussen et al., 2016). This suggests that not many and particularly larger offices are certified, which connects to the second reason for a lack of literature. It is not easy for researchers to obtain required data for these types of studies, as researchers are in lack of a centralized commercial real estate database (Kok and Jennen, 2012).

However, these problems can be overcome. Firstly, the mandatory transfer of energy labels upon project delivery, sale and lease transaction and the 2023 minimum energy label C regulation resulted in increased diffusion of EU energy labels in the Dutch office market over the past years. Secondly, Spring Real Estate, an Amsterdam based independent top full-service real estate advisory firm (PropertyNL, 2017), provided access to a real estate database that already contained a substantial pool of recent and relevant transactional and investment opportunity-related data. Moreover, Spring Real Estate provided the financial resources to retrieve the most up-to-date data to properly conduct this research. Although gathering data and information on Dutch commercial properties and investment characteristics manually is a time intensive and expensive process, it should result in reliable estimation results.

The objective of this thesis is to gain insight into an investors’ willingness to pay for sustainable office buildings in the Netherlands. This paper will aim to answer the following main research question:

1. Are investors willing to pay a premium for energy efficient office buildings?

Furthermore, a sustainability trend may be occurring in the real estate sector, i.e. that sustainability has become an increasingly important investment criterion over the past years. As will be discussed in Section 2.6, it is also important for research to examine the development of potential pricing differences between sustainable buildings over time, as standards of energy performance methods and sustainability regulations are likely to change, as is the case in the Netherlands. Therefore, this paper aims to answer the following question:

2. Has the willingness to pay for energy efficient office buildings increased over the years? As will be discussed in Section 2.4, another aspect of sustainability that is also quantifiable to a certain extent is location and transport. Several labelling schemes award sustainability credits to buildings that provide good accessibility relative to amenities and alternative transportation, such as train stations. Elaborating on the concept of sustainability, this paper will aim to answer the following research question:

3. What is the willingness to pay for the accessibility relative to amenities and train stations?

(7)

7

1.2. Research method and outline

This study will empirically examine what the effect of the level of energy efficiency and accessibility relative to alternative transport is on transaction prices for offices. Multiple regression techniques will be applied, using hedonic pricing models. Building, locational, investment and tenancy features are controlled to avoid omitted variable bias. In addition, short interviews are conducted with various real estate professionals who are asked for their opinion on the main regression results of this thesis, who are asked for their stance on the effects that the growing sustainability theme and the 2023 regulation will have on investment behaviour in the commercial real estate sector.

Preparatory to conducting any form of research, Section 2 presents a theoretical framework that elaborates on the comprehensive concept of sustainability, discusses common and available certification systems that measure the level of sustainability and elaborates on the slow adaptation of sustainability in the Dutch property market. Furthermore, Section 2 presents an overview of existing literature on the relationship between sustainability and office values and discusses their main findings and limitations. Section 3 derives the hypotheses using the knowledge acquired in the antecedent section. Section 4 firstly addresses problems of endogeneity and discusses how this paper will aim to avoid them, after which it presents the included variables and the econometric model that is used in the empirical analysis. Section 5 will discuss the data sources, present the examined sample and descriptive statistics and test the sample on the classical linear regression assumptions. Section 6 presents the obtained regression results and the conducted interviews. Furthermore, robustness tests are performed to evaluate and validate the applied methods. Section 7 concludes and positions the results in context of the earlier discussed existing literature. Finally, limitations of this paper and future research suggestions are presented in Section 8. References and appendices can be found in Section 9 and 10, respectively.

(8)

8

2. Theoretical Framework

This section provides theoretical background information on sustainability and its integration in the real estate sector. After providing various definitions of sustainability, this section provides an overview of the most commonly used sustainability frameworks used to certify buildings. Subsequently, the extent to which sustainability is integrated in the Dutch real estate market is discussed. Finally, existing empirical evidence on the relationship between the level of sustainability and building values is reviewed and discussed.

2.1. What is sustainability?

As this paper is written in the context of sustainability and aims to empirically examine the relationship between office building sustainability performance office values, it is important to define sustainability. According to Statistics Netherlands (CBS, 2017), sustainability mainly relates to scarcity of resources that generate prosperity. The surface of the earth is finite, stocks of raw material are finite in the same way as the atmosphere’s and environment’s absorption capacity are finite. Similarly, no unlimited resources of highly educated and healthy populations, well-functioning social networks, social trust, machinery and infrastructure, knowledge and other resources necessary for sustainable prosperity are available. Due to scarcity of these resources, it is not self-evident that we are currently moving towards sustainable prosperity. Therefore, sustainability policy has become a social necessity in today’s world (CBS, 2017).

Providing an unambiguous definition of the sustainability is complex, as it covers many areas in our society and economy. The most frequently cited definition of sustainability was given by the UN-commission Brundtland, formerly known as the World Commission on Environment and Development (WCED). In the “Our Common Future” report from 1987, the concept of sustainable development was described as:

"… development that meets the needs of the present without compromising the ability of future generations to meet their own needs." (Bruntland, p. 41, 1987).

With this description, the Brundtland commission established a link between economic growth, the environmental issue and both poverty and development problems.

2.2. Definitions of sustainability in the property markets

Real estate associations aiming to apply this broad and comprehensive concept to the real estate market came up with several divergent definitions.

Bak (2011), from the Dutch Association of Real Estate Brokers and Real Estate Experts (NVM), argues that it is not straightforward to state what exactly is meant by sustainability. He underlines that it is certain that sustainability goes beyond building energy-efficient properties and points out that sustainability also encompasses redevelopment of abandoned and dilapidated buildings. He discusses that, thus far, investors tend to focus too much on new construction projects rather than redevelopment, which is undesirable from a social perspective, as the greatest potential for sustainability comes from the existing stock. Moreover, he points out that the problem of the concept of sustainability is that it is often used inappropriately. Therefore, the possibility exists that necessary attention for sustainability will fade with time.

According to the Dutch Association of Institutional Real Estate Investors (IVBN, 2009), sustainability in the real estate sector means bringing together care for the environment, long lifetime, tenant satisfaction and value growth in a responsible manner. In addition, the IVBN presents their

(9)

9

perception of the term sustainable real estate, i.e. real estate that is built or modified in a specific way that minimizes scarcity of resources such as materials, energy, water and locations, while at the same time optimally functioning in terms of tenant satisfaction, indoor environment and health.

A different definition of a sustainable property is provided by Frej (2005) from the Urban Land Institute of Washington. According to her, a sustainable property, often referred to as a “green” building, is the result of a design that aims to increase the efficiency of the energy consumption, water and other raw materials, while the impact of the building on the well-being and the environment is minimized throughout the property life cycle.

According to Royal Institution of Chartered Surveyors (RICS),

“There is no universally agreed definition of a sustainable building. However, as the market develops and as new metrics are developed and regulations are implemented, a consensus may emerge. There is a general expectation that buildings that minimize environmental impact through all parts of the building life cycle and focus on improved health for their occupiers may retain value over a longer term than those that do not. Sustainable buildings should optimize utility for their owners and occupiers and the wider public, whilst minimizing the use of natural resources and presenting low environmental impact, including their impact on biodiversity.” (RICS, p. 36, 2009).

The above-mentioned definitions illustrate how complex it is to provide an unambiguous definition of a “sustainable” office building. Nevertheless, several measuring instruments are available to measure a building’s level of sustainability. Developers, investors and owner-occupiers are increasingly, either voluntarily or mandatorily, seeking to certify their buildings using one or a range of acknowledged sustainability certification systems (RICS, 2013). To gain more insight into the aspects that qualify a building as sustainable and what frameworks are available to assess sustainability performance, the most widespread used sustainability frameworks are discussed in the next section.

2.3. Sustainability Certification Systems

The previous paragraph provided some qualitative descriptions of the comprehensive concept of sustainability in the real estate market. However, it is also important to know what frameworks are available to assess sustainability performance of building, as these tools allow one to compare sustainability performance of different buildings to each other. Moreover, these methods allow one to conduct quantitative research.

Various certification or labelling frameworks are used across the world to assess sustainability performance of buildings. Winward et al. (1998) aimed to make a distinction between two types of labels: the endorsement label and the comparison label. Although both labels measure the sustainability performance of products, endorsement labels are typically voluntary and only awarded to a property in case its performance meets certain specified requirements, while comparison labels are generally mandatory2_{and of a “multi-categorical” nature. Comparison labels award properties a}

label that classifies the performance of a building on a scale from better to worse, which allows for comparison to the performance of similar objects. An example of a typical comparison label is the EU energy label. What is important to note is that, although Winward et al. (1998) aimed to make a fundamental distinction, endorsement labels for properties can also be of multi-categorical nature. Examples of endorsement labels with different ratings are BREEAM, LEED and Energy Star. Similarly, endorsement labels in the real estate sector may also contain aspects of endorsement. For example, the European performance certificate clearly ranks the energy performance of a building from good to bad, while rewarding the best performing buildings with an A-label.

2_{Mandatory transfer of a EU energy label upon sale or delivery, not only applies to the property market, but also holds for white goods, light} bulbs and automobiles.

(10)

10

The most widely applied sustainability labelling frameworks across the world and in the Netherlands will now be briefly discussed. Please note that solely frameworks that assess the sustainability performance of individual buildings are considered, i.e. not the sustainability performance of complete portfolios (e.g. GRESB).

LEED

Leadership in Energy & Environmental Design (LEED) is one among the most popular and comprehensive green building certification programs used worldwide (“Benefits of Green Building”, 2017). It was developed by the United States Green Building Council (USGBC) in 1993 with the intention to spread awareness in the real estate market, especially in the development and renovation segment. With over 32,500 certified commercial properties across the world, LEED has grown out to become the most widely applied sustainability rating system (“LEED by the numbers: 16 years of steady growth”, 2016). LEED makes distinctions between primary requirements a project must meet to qualify for a certification process, basic requirements a project must meet to qualify for a certificate and innovation points for exceptional performance beyond basic requirement. Credits are awarded based on building performance in the following main categories: management, water and energy-efficiency, indoor environmental quality, pollution, accessibility, land use, water energy-efficiency, resources and innovation (Azhar et al., 2011). Each credit is worth a certain amount of points that add up. Based on the final score, a sustainable building is awarded one of the following certificates: LEED, LEED Silver, LEED Gold and LEED Platinum (Azhar et al., 2011).

ENERGY STAR

Energy Star was developed by the U.S. Environmental Protection Agency (EPA) and Department of Energy (DOE) and launched in 1995. As opposed to other assessment tools, this instrument solely examines the intensity of energy use. Although Energy Star is widely known for assessing the energy performance of products, such as white goods, it also considers the energy performance of buildings. Moreover, it considers the performance of the entire building, rather than conducting an independent assessment of the efficiency of lights, equipment and other components. To qualify for this endorsement label, the actual energy consumption of a building is compared to a benchmark database of comparable buildings. Therefore, the label may in theory deteriorate due changing standards and improvement of the existing office stock over time. The EPA and DOE do not charge any fees for participation in the commercial buildings program. Projects that score 75 or better out of 100 points, which implies outperformance of at least 75 percent of the nationwide benchmark, are awarded with an Energy Star label (Reeder, 2010).

BREEAM

The Building Research Establishment, a British research organization, introduced the BREEAM certificate in 1990 to promote sustainable developments in the office sector (Lee and Burnett, 2008)3_.

It was the first established method that allowed for evaluation, assessment and certification of buildings. It focuses on roughly the same sustainability aspects as LEED. The assessment is carried out by recognized assessors that rate the sustainability performance of a building as “Pass”, “Good”, “Very Good”, “Excellent” or “Outstanding” by using several sustainability indices as benchmark. BREAAM has expanded its scope over the past decades and is now, not only used for developments and new buildings, but also for existing buildings and refurbishment projects. Furthermore, besides the office sector, the standard is also applied in the retail, industrial and even the residential sector.

BREEAM-NL, the Dutch spin-off, was developed by the Dutch Green Building Council (DGBC) in 2009. For this certificate to be awarded, an assessment of the sustainability performance of either in-use buildings, new-built, renovation or demolition projects is conducted by focusing on various environmentally-oriented elements, such as management, energy efficiency, indoor environment and

(11)

11

pollution. As of December 2017, BREEAM-NL covers 7.5 million square meters of floor space. The number of certificates issued has been increasing substantially over the past years. The latest recorded growth was 7.6% in the year 2017. The In-use scheme accounts for most of this growth, as its share increased from 91 in 2016 to 136 in 2017. BREEAM-NL new-construction and renovation covers 3.2 million square meters of which approximately 30% of the certificates were issued in the office-sector (Dutch Green Building Council, 2017). In 2014, “The Edge”, an office development of OVG Real Estate at the Amsterdam Zuidas, was awarded an “Outstanding” BREEAM-NL score of 98.36%, which made it the most sustainable office in the world.

GREENCALC+

Several calculation methods for sustainable performance are integrated in the Dutch Greencalc+ software tool that was launched by the Sureac Foundation. Greencalc+ can be used to test environmental ambition and performance of both individual buildings and districts. In addition to energy-efficiency, factors like used resources, water and transport are examined. Comparison of a building’s “green” performance to that of peers results in an environmental index score that qualifies a building as either sustainable or not. Different from other performance schemes, no certificate is awarded after assessment. The Sureac Foundation transferred the management and maintenance of GreenCalc+ to the Dutch Green Building Council in 2012 with the aim to create a BREEAM-NL Light version.

GPR

The GPR building instrument was developed in 2006 by the Municipality of Tilburg. This tool assesses sustainability performance of not only development projects, but also residential and commercial buildings. GPR assessors make use of recent methods and innovations that are integrated in GPR software to conduct an analysis on building characteristics. GPR focuses on building performance in five categories: energy-efficiency, environment, user quality, health and future value of the building. Performance over each category is rated on a scale from 1 to 10, with 5 corresponding to the requirements of the Dutch Building Decree (Vreenegoor et al., 2008). Like with GreenCalc+, no certificate is issued after assessment, which makes it difficult to compare the sustainability performance of buildings. These instruments serve as informative tools available to municipalities, owners and developers architects to provide insight into sustainability performance.

EUROPEAN ENERGY LABEL

In addition to the above described voluntary assessment instruments, there is also a mandatory framework that has been used as a legislative instrument by the EU striving for a more sustainable real estate sector. The European Energy Performance of Buildings Directive (EPBD) of 2006 resulted in the development of the European Energy Label in Europe. As all European member states have officially been required to implement policies that require owners to label properties upon delivery, sale or lease since 2009 (Bio Intelligence Service et al., 2013), the energy label has grown out to become the most commonly applied sustainability framework across Europe. However, as the name suggests, only the energy efficiency aspect of sustainability is assessed. Furthermore, it is important to note that this certificate only takes theoretical building-related energy consumption into account, the actual energy consumption of the building is therefore not considered. Because buildings could also generate energy instead of solely consume it, by using solar panels for example, the assessment focuses on net rather than gross energy consumption. An energy index score measures the theoretical level of energy efficiency of a property. The energy index score for utility buildings such as office properties is calculated as follows:

(12)

12 𝐸𝑛𝑒𝑟𝑔𝑦 𝐼𝑛𝑑𝑒𝑥 =𝑄𝑝𝑟𝑖𝑚;𝑡𝑜𝑡;𝑏𝑢𝑖𝑙𝑑𝑖𝑛𝑔 𝑄𝑝𝑟𝑖𝑚;𝑡𝑜𝑡 𝑝𝑒𝑟𝑚 Where: 𝑄𝑝𝑟𝑖𝑚;𝑡𝑜𝑡 𝑏𝑢𝑖𝑙𝑑𝑖𝑛𝑔 = 𝑇𝑜𝑡𝑎𝑙 𝑒𝑛𝑒𝑟𝑔𝑦 𝑐𝑜𝑛𝑠𝑢𝑚𝑝𝑡𝑖𝑜𝑛 𝑏𝑢𝑖𝑙𝑑𝑖𝑛𝑔 (in MJ) 𝑄𝑝𝑟𝑖𝑚;𝑡𝑜𝑡 𝑝𝑒𝑟𝑚 = 𝑇𝑜𝑡𝑎𝑙 𝑝𝑒𝑟𝑚𝑖𝑡𝑡𝑒𝑑 𝑒𝑛𝑒𝑟𝑔𝑦 𝑐𝑜𝑛𝑠𝑢𝑚𝑝𝑡𝑖𝑜𝑛 𝑏𝑢𝑖𝑙𝑑𝑖𝑛𝑔 (in MJ) Source: Isso (2013)

The exact guidelines and more extensive formula that should be used to calculate the EPC-index are presented in the NEN 7120 – Energy performance of buildings – Determination method (“Bepalingsmethode EPC”, 2018). In an EU Energy Performance Certificate (EPC), both an energy index score and the corresponding energy label category are issued (ranging from categories A++ to G). As illustrated in Table 1, the relationship between the energy index and energy labels is non-linear, as the index-spread distribution is non-constant. The clear ranking of this assessment tool makes it a typical example of a comparison label.

Table 1: Energy index spreads and corresponding energy labels

This table presents the different index-based energy label categories. Buildings labelled with an energy label ranging from A++ to C, will be referred to as energy efficient rather than sustainable. Similarly, buildings with an index above 1.30 will be referred to as energy inefficient, rather than non-sustainable.

Energy Label A++ A+ A B C D E F G

Energy Index 0.00-0.51 0.51-0.70 0.71-1.05 1.06-1.15 1.16-1.30 1.31-1.45 1.46-1.60 1.61-1.75 >1.76 Source: Isso (2013)

Approximately 3.2 million energy labels have thus far been issued in the Dutch residential sector (“Energielabels van woningen”, 2017). Arnoldussen et al. (2016), estimated that only approximately 23% of the Dutch office stock - the equivalent of 18.8 million square meters of leasable floor area – are provided with an EPC, while this corresponds to only 6% of the total number of offices (approximately 4500 offices) (Arnoldussen et al., 2016). This disproportionality suggests that not many and particularly larger offices are certified. It is important to point out that these estimates correspond to the year 2014. Stricter enforcement of the regulation that requires energy labels to be transferred upon project delivery, sale and lease contracts, and the 2023 minimum energy label C regulation are likely to have resulted in greater diffusion of EU energy labels in the Dutch office market over time.

2.4. Sustainability revisited

As illustrated by the previous sections, there are numerous methods in the global property market to determine whether a building can be considered sustainable. Besides using several descriptions of what makes an office building sustainable, several tools are available that allow for quantitative assessment of the level of sustainability. The diverse descriptions and labelling frameworks confirm that it is difficult to unambiguously determine what qualifies a building as sustainable. There is no one definition (RICS, 2009) nor is there one classification used across the world.

Although the definitions provided in Section 2.2 to some extent clarify what is meant by sustainability in the real estate sector, definitions are difficult to quantify and therefore make it hard to rate the level of sustainability of a building and compare characteristics of different buildings to each other. Consequently, it would be more informative to appeal to sustainability instruments for studies that examine the environmental performance of buildings. Table 2 provides an overview of the assessment tools that were earlier discussed, together with the assessment criteria considered. These are the most common instruments used worldwide and in the Netherlands. As can be derived from Table 2, LEED and BREEAM are the most extensive instruments that allow for assessment of the level of sustainability. Therefore, they offer the most comprehensive classification of sustainability in

(13)

13

real estate. Ideally, all office buildings should be therefore either BREEAM or LEED-certified. In practice however, many costs are associated with these certification methods (Aspinall et al., 2012; Salminen et al., 2012), which is the main reason that he Dutch property market has, thus far, been hampered by the slow dispersion of divergent sustainability performance certification frameworks (Kok and Jennen, 2012).

Table 2: Comparison of assessment criteria of sustainability certification

systems

This table presents the aspects and compares the criteria of the most commonly used certification methods to assess the level of sustainability of office buildings. What strikes is that these approaches all address a different set of elements of sustainability.

LEED Energy Star BREEAM (-NL) GreenCalc+ GPR Energy Label

Management ✓ - ✓ - - -

Energy Efficiency ✓ ✓ ✓ ✓ ✓ ✓

Health ✓ - ✓ - ✓ -

Pollution ✓ - ✓ - - -

Location & Transport ✓ - ✓ - - -

Land use & Ecology ✓ - ✓ - - -

Water ✓ - ✓ ✓ ✓ -

Resources ✓ - ✓ ✓ ✓ -

Innovation ✓ - ✓ - ✓ -

Certification expenses High Free High Average Low Low

Source: Own research

This raises the question which workable and quantifiable classification can be used in the Netherlands to designate sustainable office buildings. Although it solely focuses on energy consumption and only approximately a quarter of the Dutch office stock was provided with an EU energy label in 2014, it is the most workable tool to indicate whether an office building is “sustainable” or not. This explains why the Dutch government has, thus far, used the EU energy label as one of its main policy instruments to make sustainability-related decisions in the property market, such as the regulation of mandatory transfer of an energy label upon project delivery or in case of sale and lease transactions and of course the announced regulation that requires all functioning office buildings to be certified with a minimum energy label C as of January 1, 2023.

Because the energy performance certificate only considers the energy consumption aspect of buildings, it is clearly not the most comprehensive measure of sustainability performance. Therefore, it would be of added value to also consider other quantifiable sustainability assessment criteria. It can be derived from Table 2 that various aspects play a role. However, not all are easy to obtain data on, except for one: location and transport. Within this category, credits are rewarded for various location and transportation-related criteria by both LEED and BREEAM(-NL). For example, for the accessibility relative to amenities and alternative transportation, such as train stations. As this study aims to model the effect of office sustainability on transaction prices and as it is possible to gather data on the latter two sustainability-criteria as well, they will also be incorporated into this thesis.

To avoid confusion throughout this paper, the research questions focus on the concepts of energy efficiency and accessibility separately rather than solely on the concept of sustainability.

(14)

14

2.5. Implementation of sustainability in the Dutch office market

The previous paragraph elaborated on the concept of sustainability, explained which factors play a role in sustainable building performance and which certification methods can be used. This paragraph discusses the slow adaptation of sustainability in the Dutch property market.

Investors, developers, housing associations, companies and tenants are increasingly encouraged to take sustainability measures, not only because of increasing social and environmental pressure, but also because of stricter policies and regulations.

As of January 1, 2008, it is compulsory for owners of existing residential and utility buildings in The Netherlands to transfer an energy label in case of a sale or rental transaction. Furthermore, every completed new-built office project should be labelled as of July 1st_{, 2014 (Lenteakkoord, 2014).}

However, because an effective sanction had been missing up until recently, not every building was being labelled. As of the beginning of 2015 however, the Dutch Inspection on Living Environment and Transport was enabled by the government to fine owners in absence of an energy label. The maximum penalty for legal entities amounts to €20,250. The largest fine was issued in 2017 and amounted to more than €10,000 (“Energielabel utiliteitsgebouwen”, 2018). As of 2020, all new-built construction permit applications, both for residential and non-residential construction, must comply with the BENG-requirements4_{. This new regulation resulted from the Energy Agreement for sustainable growth}

and the European directive EPBD. At the Climate Summit 2015, 175 countries signed an international climate treaty with the aim to limit global warming and improve the global climate. To allow for suitable implementation of these agreements in the Netherlands, a national climate summit was organized in October 2016. One of the outcomes of the summit was that commercial properties in The Netherlands with an EPC of D-G may no longer function as office buildings as of 2023 (Ministerie van Binnenlandse Zaken en Koninkrijksrelaties, 2016).

On April 1, 2016, the Dutch Green Projects Scheme came into force. This program stimulates environmental friendly investments. Financiers receive tax benefits, which results in lower interest rates of so-called “green loans”. In addition, the national government itself also contributes to the sustainability of the Dutch property stock by only purchasing or leasing buildings that are certified with an energy label C or better.

Besides stricter sustainability legislation and evident benefits for society, several other factors can be identified that should drive the implementation of sustainability in the real estate sector. Both owners and tenants can benefit from owning and leasing more sustainable buildings (Warren et al., 2009; Jones Lang LaSalle, 2007). On the one hand, owners benefit from enhanced reputation, increased revenues, higher lessee quality, increased rent levels, a higher chance of meeting new future regulations, less vacancy, lower operational expenses and more stakeholder satisfaction. On the other hand, tenants profit from enhanced reputation, lower operating costs, improved productivity, enhanced building quality, more attraction of employees and a stronger relationship between tenant and owner.

Despite social pressure, tightening laws and regulations and theoretical advantages, sustainability has still only been slowly implemented in the property market. A discrepancy seems to exist between the theoretical benefits of “green” labels and their slow diffusion in the market (Kok et al, 2011; Kok and Jennen, 2012; Eichholtz, 2010). Real Estate actors have responded slowly to sustainability trends. Cadman (2000) provides an explanation for the slow transition to sustainable real estate by introducing the “vicious circle of blame” (Figure 2.5).

4_{BENG stands for ‘bijna energieneutrale gebouwen’ which translates into ‘almost energyneutral buildings’. The requirements can be} retrieved from: https://www.rvo.nl/onderwerpen/duurzaam-ondernemen/gebouwen/wetten-en-regels-gebouwen/nieuwbouw /energieprestatie-beng/wettelijke-eisen-beng

(15)

15 Figure 2.5: The vicious circle of blame

Source: Cadman (2000)

Figure 2.5 presents four stakeholders that are both able and willing to contribute to a more sustainable real estate market. However, they do not believe other stakeholders are willing to contribute as well. Occupiers, constructors, developers and investors blame each other for either not demanding or supplying sustainable buildings. Lorenz (2008), expanded the original Cadman model by introducing parties, such as policy makers, researchers, insurers and financiers. It is of social and environmental importance for the circle to be broken. Sedlacek and Maier (2012) suggest that, although government regulations can serve as a useful instrument, the dilemma can most adequately be solved by providing market participants with more information about the performance of sustainable real estate. Conducting research and spreading more awareness among stakeholders will lead to faster diffusion and adoption of sustainability in the property market.

A lot has changed in the real estate market since the introduction of the vicious circle of blame in 2000. The demand for sustainable buildings has increased, because of treaties, directives, policies, regulations and social pressure. Moreover, an increasing amount of information on performance of sustainable real estate is available because of academic research. As this paper aims to model the relationship between sustainability characteristics and office value, it is important to read into existing research that has been conducted in this field. Therefore, the next section provides an overview of relevant existing literature.

2.6. Literature Review

In this section, existing and relevant literature on the relationship between sustainability and building values is reviewed. Appendix 1 presents an overview of the main findings of the research papers that are discussed below, together with their segment of interest, sample information, used data source, regression method, sustainability measure, dependent variable and found premiums.

This is not the first paper to examine the relationship between environmental performance and building value. Sayce et al. (2010) discuss that, although outcomes of interviews typically suggest that investors pay extra for more sustainable properties, there is a need for quantitative research as well, as only empirical examination can provide true evidence on the effect of building sustainability and value. Most of the literature reviewed in this section focuses on existing papers that link data on

(16)

16

sustainability with data on either property or rental values. As Dutch empirical evidence on this topic is scarce, this paragraph mainly focuses on foreign literature. Foreign related literature is found to focus predominantly on the American office market. According to Mudgal et al. (2013), an explanation for this is the early adoption of two labelling frameworks in the U.S. that were discussed in Section 2.3: LEED and Energy Star. Furthermore, transactional data and data on office characteristics can be easily retrieved from the U.S. CoStar-database (Kok and Jennen, 2012).

Foreign existing literature

Miller et al. (2008) examine whether LEED and Energy star certified prime U.S. offices trade at a premium. They examine this by comparing a certified sample to a peer group that is not certified. The writers highlight the potential problems of omitted variable bias. For example, they reason that certified buildings are likely to be newer. To mitigate endogeneity problems, they control for property characteristics like age, location and size. They find that offices with LEED and Energy Star certificates trade for a premium of 9.9% and 5.8% respectively, statistically significant at a 10% level. According to Reichardt et al. (2012), a large shortcoming of the paper of Miller et al. (2008) is that they do not account for micro-locational effects when comparing the “green” to “non-green” samples. This could result in a bias if non-certified offices are generally located in less favourable regions of submarkets with respect to for example accessibility, compared to labelled offices that could be structurally located superiorly. Therefore, Reichardt et al. (2012), who are interested in rental premiums, apply difference-in-difference (DID) and Fixed Effects (FE) regression techniques using two observations for each building: both pre and post certification. This allows them to control for unobserved factors and compare labelled and non-labelled offices within the same submarket. The writers find significant rental premiums for both LEED and Energy Star labelled buildings of respectively 2.9% and 3.5%. In addition, Reichardt et al. (2012) track the development the Energy Star rental premium over time. They do this by examining a dataset of office buildings with at least two rental transaction per object: one prior to the issuance of the sustainability certificate and one after. Naturally, this very accurately controls for dissimilarities between the control and treatment group. After finding the green rental premium to be non-constant over time, partially because of the global financial crisis, they emphasize that it is of importance for future studies to examine the development of the difference in pricing, as standards and guidelines of labelling methods evolve and office supply could change because of stricter regulations. The latter is one of the main motivations for writing this thesis.

In the first of their series of papers, Fuerst and McAllister (2009a) compare LEED and Energy star labelled offices to non-labelled offices. Non-labelled properties are selected from the same sub-market as labelled buildings. In line with Reichardt et al. (2012), they control for micro-location and property specific effects to mitigate problems of unobserved heterogeneity. They find that LEED and Energy Star-labelled buildings trade for, respectively, a 31% and 35% premium and are found to transact at 6% higher rental levels. Although their sample period is unspecified, they point out that it is of their concern that existing literature predominantly tends to examine short time periods. This could be problematic because, as supported by the findings of Reichardt et al. (2012), pricing effects could change over time. Furthermore, they discuss that future work should also focus on the effect of labelling in other commercial sectors, as this may lead to different results.

This is exactly what Fuerst and McAllister (2011a) did in a follow-up study. Examining not only U.K. offices, but also industrial and retail properties, they do not find any strong evidence of market and rental values to be driven by energy performance. They conclude that EU Energy Performance Certificates do not financially incentivize market actors to carry out energy efficient investments for commercial property assets. Fuerst and McAllister (2011a) controlled for building characteristics, however they advisedly do not control for micro-locational differences, since they have no reason to believe that energy efficient buildings in their sample are structurally located more favourably within regions than non-labelled properties. In another paper, Fuerst and McAllister (2011b), focused on the U.S. office market again. Different from other studies, they address the importance to control for the number of tenants, because of potentially structural differences in the ratio of single-tenant and

(17)

multi-17

tenant offices between the labelled and non-labelled groups. They find LEED and Energy Star labelled offices to, respectively, transact at premiums of 25% and 26%, while tenants are only found to pay rental premiums of 5% and 4% for these office buildings. In a follow-up study, Fuerst and McAllister (2011c) use a larger and more up-to-date sample. After controlling for robustness to address the impact of outliers, they find that Energy Star-labelled offices to transact at a premium of 18%, while finding similar results to their prior study, relative to both rental and transactional premiums for LEED-labelled offices and rental premiums for Energy Star-certified offices. In line with prior work (i.e. Fuerst and McAllister, 2009b), they find labelled office buildings to, on average, have higher occupancy rates. In similar work, Eichholtz et al. (2010) examine a pre-crisis office building sample and apply hedonic techniques to estimate the effect of eco-labelling on transaction and rental value. After controlling for a set of economic and hedonic characteristics, they find both premiums for LEED certificates to be insignificant. However, they do find significant transactional and rental premiums of 19.1% and 3.3% for Energy Star labelled offices. To control for locational quality differences of submarkets, they compare labelled to non-labelled offices within a 0.2 square miles geographical area. Fuerst and McAllister (2011b) argue that this is not an accurate proxy for submarkets, as locational quality of buildings and district-density may differ substantially between submarkets. Despite this comment, Eichholtz et al. (2013) still use the 0.2 radius in a follow-up study, although controlling more rigorously for differences in characteristics between labelled and non-labelled buildings. In addition to their previous work, they are interested in whether their findings also hold during the crisis. They find significant transactional and rental premiums for LEED, as well as Energy Star certified offices, of respectively 11.1% and 6.0% for LEED and 12.9% and 6.5% for Energy Star. Furthermore, Eichholtz et al. (2013) find more sustainable office buildings to be of higher building quality than non-sustainable buildings, as the group of green buildings consists of 75% “Class A” offices, while the ‘non-green’ buildings group consists of only 26% “Class A” offices. Moreover, they find sustainable office buildings to be younger, larger, located more favourably with respect to transport and facility, which emphasizes the importance of controlling for unobserved heterogeneity. Different from other studies, Eichholtz et al. (2013) address the “vintage” effect of the Energy Label certificate, as they find that the willingness to pay for Energy Star labelled building yearly decreases with 0.4%, measured from the day it was issued. This finding may be explained by changing labelling standards.

Wiley et al. (2010) examine the effect of labelling on value of prime offices using CoStar data like all other U.S. studies. Although the examined period is unspecified, they find LEED as well as Energy Star certified offices to trade for premiums of, respectively, $130 and $30 per square foot. Furthermore, they find that rental premiums ranging from 15-18% and 7-9%, respectively. Fuerst and McAllister (2011b) note that Wiley et al. (2010), like other papers5_{, could have controlled more}

accurately for location quality differences within subareas because they expect sustainable office buildings in the sample to be structurally located more favourably than non-labelled ones.

Although most international is conducted in the U.S., there are more European studies besides the earlier discussed U.K. study of Fuerst and McAllister (2011a). Interested the relationship between European EPC-ratings and office values in Sweden, Bonde and Song (2013) use hedonic techniques and find energy performance not to be capitalized in appraisal values of office buildings. They address several important limitations to their own paper. Firstly, they rely on appraisals rather than transactional data, which could pose complications as appraisers tend to smooth valuations over time. Moreover, Bonde and Song (2013) argue that it may be complicated for appraisers to account for asset heterogeneity. Therefore, they suggest future research to focus on transactional data instead of appraisal values, as that would provide more robust evidence and reveal the true willingness to pay for energy efficient commercial properties. Finally, they discuss that not having accounted for lease structure of the examined buildings, which was not possible due to confidentiality reasons, could have resulted in an omitted variable bias.

In similar work, Chegut et al. (2013), contribute to existing literature in several ways. They examine whether U.K. BREEAM-labelled offices are associated with a positive effect on both capital

(18)

18

and rental values, using transactional data rather than appraised values. They contribute, not only by controlling for location and building characteristics, but also for the effect of investor characteristics and lease structure. Chegut et al. (2013) find all categories to be of influence on the effect of interest. They find both significant rental and price premiums for labelled offices compared to non-labelled offices of, respectively, 26% and 21%. They recommend future research to account for investor characteristics, as they find these to be of modifying impact on the effect of interest. An evident limitation of their study is the small amount of observations6_.

In another Swedish study, Cerin et al. (2014) want to examine whether energy efficient properties transact at a premium, since the EU directive that made Energy Performance Certificates mandatory for all residential properties was implemented. As residentials are considered relatively homogeneous objects relative to offices, they only control for a limited number of hedonic characteristics. Cerin et al. (2014) find energy efficient houses to transact at a significant premium of 6 percent.

Dutch existing literature

As opposed to the extensive empirical evidence from the U.S., existing literature on the impact of sustainability on office values in the Dutch commercial real estate sector is rather scarce. This is confirmed by Brounen and Kok (2011), who underline that their paper contributes, as most existing literature predominantly focuses on the U.S. commercial real estate sector. Brounen and Kok (2011) empirically examine whether Dutch residential properties with a “green” energy label transact at a premium. Their findings suggest that dwellings with an EU energy label ranging from A to C trade at an average premium of 3.7% compared to properties with energy labels ranging from D to G. These findings are interesting, as the adoption rate of energy labelling was low and declining at the time the study was performed. Brounen and Kok (2011) address the need for stricter energy certification policies to improve the energy efficiency level of the overall residential stock. Within the same year, the EU parliament announced new legislations that eliminated all waiver-options for homeowners and made energy performance certification strictly mandatory at time of transaction.

Kok and Jennen (2012) are interested in a similar effect for offices. They address the difficulty of gathering transactional data on commercial real estate in the Netherlands, as researchers are in lack of a centralized database, such as CoStar. They overcome this problem by merging databases of various Dutch brokerage firms, which allows them to examine whether offices with “green” energy labels earn rental price premiums in the Dutch office market. Although the examined sample is arguably outdated7_{and the writers focus on rental data, their findings are very relevant to this thesis,}

as this is the only Dutch academic paper that focuses on realized price data rather than appraised values. Although the samples differ, results of their study and this thesis could provide the first indicative insights into the financial feasibility of owning more energy efficient offices. Kok and Jennen (2012), unlike this thesis, do not extensively address the complexity of the ambiguous and often misunderstood concept of sustainability, even though it can be derived from Section 2.1 – 2.4 that a lot of factors play a role in assessing the level of sustainability, Kok and Jennen (2012) argue that the most important components are energy efficiency and accessibility. What strikes is that they measure the impact of accessibility on prices not only by examining the effect of distance relative to amenities and nearest train station, but remarkably also to the nearest highway ramp. However, as was stated in Section 2.4, the most comprehensive schemes reward sustainability points for the accessibility relative to amenities and alternative transportation, not the nearest highway junction. In fact, LEED and BREEAM-NL actually discourage car-use by rewarding credits to sites that stimulate the use of green vehicles and that either do not exceed minimum parking requirements or that “outperform” the parking ratio of benchmarks. Despite this theoretical limitation, their findings suggest that office properties with a label ranging from D to G are leased for a discount of 6.5% compared to similar, but

6_{Chegut et al. (2014) compare 64 certified leases and 68 certified transactions to respectively 1149 and 2103 non-certified leases and} transactions.

(19)

19

less energy efficient office buildings. In line with the findings of Eichholtz et al. (2013), they find accessibility relative to train stations and nearby or on-site amenities to be priced in as well8_. In line

with Reichardt et al. (2012), they also argue that it is important to track the effect of interest over time, not only because of changing standards, but also because macro-economic conditions that may affect the results. For instance, Kok and Jennen (2012) find realized rents of green offices to rise more under more favourable macro-economic conditions, while falling harder in times of economic downturn.

Rather than focusing on rental prices, former brokerage firm DTZ (2011) examined the effect of energy efficiency on appraised office values. They find the effect only to be evident for top-locations within the Randstad, which is supported by similar work by Heineke (2009). DTZ finds the energy index of offices to significantly impact appraisal values. The premium more energy efficient buildings earn is found to become stronger as the index increases. As discussed, however, it is important to note that appraisal values can suffer from smoothing and lagging and therefore may not be representative for the true market value (Clayton et al., 2001). Examination of the effect of energy efficiency on transaction prices would therefore be more informative (Bonde and Song, 2013; Chegut et al., 2013), which is exactly what consultancy firm Troostwijk Real Estate (2011) aimed to do. The writers found office price and rental price premiums of respectively 30% and 25%. However, what strikes is that Troostwijk Real Estate, unlike any other study, does not correct for any crucial determinants of office value, such as building year and location. Kok and Jennen (2012) also point out this obvious limitation of Troostwijk’s study. Accordingly, their findings present rather biased associations than causal relationships. As illustrated by the findings of Kok and Jennen (2012), the green rental premium decreases substantially after having controlled for various hedonic characteristics. Moreover, it is also important to point out that it is debatable to what extent the outcomes of DTZ (2011) and Troostwjjk Real Estate (2011) can be considered reliable, as brokers and real estate consultancies may be incentivized to carry out news that does not conflict with the global consensus.

It can be derived from Appendix 1 that most of the existing literature is relatively outdated and mainly examines short time periods. Furthermore, most existing academic literature focuses on the U.S. commercial real estate sector, as data can be easily retrieved from centralized databases. Outcomes of existing literature seem relatively consistent: a higher level of sustainability results in higher capital values and higher rental values. From the literature discussion above, it follows that existing literature recurrently addresses the importance of accounting for unobserved heterogeneity, as not correcting for building, locational, lease structure and investor characteristics could lead to problems of endogeneity. The importance of controlling for these characteristics is illustrated by the fact that, although every U.S. study makes use of the same database and relatively similar samples, divergent magnitudes are found for the same effect of interest.

The little existing European and Dutch literature focuses primarily on the impact of energy efficiency on either rental prices or appraised office values, even though use of transactional data would lead to more robust evidence (Bonde and Song, 2013; Clayton et al., 2013). Furthermore, Dutch empirical evidence is noticeably outdated, examines mainly short time periods and small samples. Moreover, no Dutch research on the relationship between sustainability and office value has been conducted since the 2023 minimum energy label C requirement was announced, while it could have serious consequences for the Dutch office stock (EIB, 2016)9_{. This is supported by Reichardt et al.}

(2012) and Kok and Jennen (2012) who, as discussed, address that future research should continue to examine the development the pricing difference for sustainable buildings over time, as standards and

8 _{DTZ Zadelhoff (2011), Chegut et al. (2014) and Fuerst and McAllister (2011c) found insignificant premiums for train station accessibility.} Fuerst and McAllister (2011c), Eichholtz et al. (2010) and Eichholtz et al. (2013) found premiums for the presence of nearby or on-site amenities.

9_{The Economical Building Institution estimated that 0.9% of the office stock will disappear as a result of the 2023 minimum label C} requirement, which corresponds to 720,000 square meters of leasable floor area. Investors of these offices are not able to make the required investment, as the relatively high investment costs outweigh the relatively low rent levels. 70% of the approximated to be withdrawn office stock concerns office buildings with an energy label G and 30% are offices labelled with an energy performance of D-F.

(20)

20

guidelines of labelling methods evolve and the supply may change because of stricter regulations or macro-economic conditions.

Consequently, it can be stated that there is a shortage of empirical evidence on the relationship between sustainability and transaction prices in the Dutch office market. The lack of literature can be explained. Firstly, the European real estate property market has been hampered by lingering dispersion of divergent sustainability performance certification schemes (Brounen and Kok, 2010; Kok and Jennen, 2012). By 2014, only approximately 23% of the Dutch office stock was provided with an Energy Performance Certificate (Arnoldussen et al., 2016), which is the equivalent of 18.8 million square meters of leasable office floor area. However, this percentage corresponds only to a mere 6% of the total number of offices, which suggests that particularly large offices are certified (Arnoldussen et al., 2016). Besides the evident lack of certified offices, gathering data on recent transaction prices and property characteristics is an expensive and difficult process (Kok and Jennen, 2012). Dutch researchers are in lack of a centralized database, such as the American CoStar database that contains transactional prices and characteristics, which prevents researchers from the ability to empirically examine whether energy efficient office buildings trade at a transactional premium. The aim of this study is to close this existing research gap in The Netherlands.

(21)

21

3. Hypotheses

In this section the hypotheses will be derived from the previous section.

This research aims to provide empirical evidence on whether investors are willing to pay a premium for office buildings with “green” energy performance certificates. As discussed in Section 2.6, most existing literature found sustainable and energy efficient properties to either trade for or to be appraised with a premium, compared to similar and either non-labelled less energy efficient buildings. Consequently, the following null-hypothesis and alternative hypothesis are formed:

Hypotheses research question 1

(𝐻₀): Investors are willing to pay as much for energy efficient office buildings as they are for similar energy inefficient office buildings.

(𝐻1): Investors are willing to pay a premium for energy efficient office building compared to similar energy inefficient office buildings.

It can be derived from the discussion in Section 2.4, that another aspect of sustainability is also quantifiable, namely location and transport in the form of accessibility relative to amenities and alternative transportation, such as train stations. Moreover, existing literature recurrently addresses the importance to control for micro-location differences in addition to controlling for submarkets. The main reason for this is that energy efficient office buildings may structurally be located more favourably compared to energy inefficient buildings, which could lead to misattributions to the effect of interest. Controlling for accessibility is the main way of controlling for the favourability of office locations within a given subarea. Accessibility will be defined as attractiveness of a location relative to neighbourhood amenities and facilities, the nearest train station – taking the station quality into account – and the nearest highway junction. As discussed in Section 2.6, sustainable aspects of accessibility are distance to nearby or on-site amenities and the nearest train station. As Kok and Jennen (2012) found rental prices to be positively driven by these factors, the following null-hypothesis and alternative null-hypothesis are formed:

(𝐻0): Investors are willing to pay as much for offices that are characterized by more favourable accessibility relative to amenities and train stations as they are for otherwise similar office buildings.

(𝐻₁): Investors are willing to pay a premium for offices that are characterized by more favourable accessibility relative to amenities and train stations compared to otherwise similar office buildings.

Reichardt et al. (2012) and Kok and Jennen (2012) found that is possible for energy efficiency premiums to change over time and address the importance to examine the development in case of stricter certification standards and tighter regulations, which was supported by findings of and Eichholtz et al. (2013). An example of such a regulation is the regulation that was announced in October 2016. This regulation will, as of January 1, 2023, require all functioning offices to be labelled with at least an energy label C. This, together with growing attention society has been paying to sustainability, raises the question whether energy efficiency has become an increasingly important investment criterion over the examined period and leads to the following hypotheses:

(𝐻0): Over the past years, investors have been willing to pay a relatively constant premium for energy efficient office buildings compared to similar energy inefficient office buildings. (𝐻1): Over the years, investors have been willing to pay an increasing premium for energy efficient

The willingness to pay for energy efficient and accessible office buildings