The impact of renewable on-site energy production on property values

2.1 Drivers for property investors to invest responsibly

The drivers influencing green building adoption can be grouped into the following categories: external drivers, corporate and individual level drivers and project- and property-level drivers (Darko et al., 2017). External drivers, such as governmental and international regulations, policies and incentives, demand and pressure from clients, rising awareness and the number of tools to implement sustainability, have proven to be important drivers for companies to act sustainably (Sayce et al., 2007; Braun et al., 2017; Nousiainen and Junnila, 2008). Sustainability is also a way to enhance corporate image and increase differentiation from competitors (Andelin et al., 2015). Marketing benefits strongly encourage sustainability: green signalling is an important means to position properties better in the eyes of tenants and potential buyers. Because people are the ones making decisions within organisations, it is also important to understand how evolving social consciousness drives individuals to implement green building practices in their projects and organisations (Darko et al., 2017; Braun et al., 2017).

In addition to external drivers, corporate- and individual-level drivers, benefits both at the project (e.g. construction costs and time, safety, reduced amount of waste, etc.) and property level are crucial, the latter of which is the focus of this article. Falkenbach et al. (2010) present property-level drivers as components that influence cash flows and, thus, the value of properties. Directly adjusting single valuation input parameters is one way of considering sustainability in valuations, with the other methods being a lump-sum adjustment or calculation of a sustainability correction factor (Lorenz and Lützkendorf, 2011). Enhanced sustainability may have the following effects on the valuation parameters (RICS, 2013; Lorenz and Lützkendorf, 2011; Lorenz and Lützkendorf, 2008; Lützkendorf and Lorenz, 2005): increased rent levels, rental growth and occupancy rates decreased operating expenses and reduced risk level (yield), which is usually based on reduced risk of obsolescence and regulatory risk.

2.2 Empirical evidence: green certificates

Green certificates and energy efficiency labels (hereafter certificates) have drawn much attention in empirical sustainable property investment research; these studies have indicated a clear relationship between sustainability investments and better financial performance. With reference to the above property-level drivers, properties with green certificates have been reported as having higher rents (Fuerst and McAllister, 2009b; Fuerst and McAllister, 2011; Reichardt et al., 2012; Reichardt, 2014; Holtermans and Kok, 2019), as well as a lower vacancy (Miller et al., 2008; Fuerst and McAllister, 2009a; Wiley et al., 2010; Reichardt et al., 2012) and operating costs (Newsham et al., 2009; Pivo and Fisher, 2008; Reichardt, 2014). These enhancements, together with the green signalling effects of certificates (Fuerst et al., 2016; Reichardt, 2014), lead to better sales prices of certified properties. Sales price premiums (Pivo and Fisher, 2008; Fuerst and McAllister, 2009b; Eichholtz et al., 2010; Fuerst and McAllister, 2011; Eichholtz et al., 2013; Chegut et al., 2014; Kahn and Kok, 2014; Fuerst and McAllister, 2009b; Eichholtz et al., 2010; Fuerst and McAllister, 2011; Eichholtz et al., 2013; Chegut et al., 2014) are typically found to be much larger than the enhancements in cash flow parameters. This might be because of the combined effects on the capital value of better rental income, lower vacancy and operating expenses, as well as lower risks (yield).

Beside property-level indicators, corporate-level benefits, especially higher returns and lower risks, have gained some attention in the earlier research. Listed green real estate funds and companies – measured, for instance, by the number of certified assets or other green scoring systems – offer possibilities for these inspections. For instance, Sah et al. (2013) report that green real estate investment trusts had 5.68% higher annualised returns compared with their less-green peers. Fuerst (2015) observe that green real estate investment trusts in North America, Europe and Asia had higher asset and equity returns, as well as higher returns per unit of risk. Cajias et al. (2012) state that among the listed European real estate companies, sustainability increased companies’ ability to generate revenues and decreased the level of idiosyncratic stock volatility. Further, Eichholtz et al. (2012) note that green real estate investment trusts have lower systematic risk, suggesting that green portfolios are better protected against certain risks, such as rising energy prices and changing environmental legislation. An and Pivo (2017) concentrate on default risks and report that green buildings in commercial mortgage-backed security loans had a 34% lower default risk than regular buildings. Lower risks have also been reported on the asset level in the form of decreased yields (McGrath, 2012; Miller et al., 2008).

Green buildings may also enjoy better loan conditions (Eichholtz et al., 2015; An and Pivo, 2017), improving the net cash flow of properties from investors’ perspective. Besides these financial indicators, Devine and Kok (2015) document the intangible benefits as follows: higher tenant satisfaction, a higher probability of renewing leases and decreased tenant rent concessions in certified buildings compared with non-certified ones. However, the tangible and intangible benefits must outweigh the additional cost for the investment to be profitable. Evidence of this is very sparse. Further, the research is usually based on a few case studies, making generalisation of the results difficult. Some of the existing studies found green building cost premiums of up to 20% (Dwaikat and Ali, 2016; Kats, 2010), and some studies have found no statistical difference between conventional and green buildings (Matthiessen and Morris, 2007; Rehm and Ade, 2013). A study by Chegut et al. (2019) is the first to compare both the green building cost premium and benefits on a large UK data set; they report higher marginal construction costs for green buildings but even larger benefits. However, higher design costs and longer construction time of green buildings compared with conventional ones increases developers’ risks, hindering developers from adopting green features in buildings.

2.3 Empirical evidence: renewable on-site energy production

Another track of sustainable property investment research has delved into the effect of PVs on house sale prices, the vast majority of those using statistical analysis in different states in the USA, especially California. These studies have been able to find a positive relation between house values and installed PVs. Hoen et al. (2017, 2013) find strong evidence that houses with PV have sold at a premium over comparable houses without PVs. Based on their data set of 72,000 houses in California sold between 2000 and 2009, the writers find a premium of $3.9-6.4 per installed watt in those 2,000 houses with PVs, with most models coalescing at $5.5 per watt. The installation cost of PV systems amounted to $5 per watt within the same time period in California, implying that PVs would be profitable investments. When the data set was split among new and existing houses, the writers show that existing houses had higher premiums and that when PV systems age, the premium decreases. Using a data set of approximately 490,000 transactions in their statistical analysis and approximately 80,000 transactions in their repeat sales analysis, Dastrup et al. (2012) find a premium of 3-4% for houses with PVs over comparable houses without PVs in San Diego and Sacramento, CA. Combining demographic data into their house sales data, they also show that the sales price premium varies with neighbourhood characteristics such as environmental ideology, income and education level. One of the earliest studies included Farhar and Coburn’s (2008) repeat sales analysis concerning 15 energy-efficient houses in San Diego, where they provide evidence of higher sales prices in PV houses over 12 comparable houses.

In addition to the Californian studies, other US markets have been analysed. Using house resale and PV building permit data from 2000 to 2013 in Hawaii, Wee (2016) found that the presence of PVs adds an average of 5.4% to the value of a house. Based on transaction and valuation data in Arizona, Qiu et al. (2017) conclude that houses with PVs enjoy a premium of 15% compared with the median house value and a premium of 17% over median house sales price in the control groups; their result suggests that PV investment costs are offset by the value increase, at least on average. In the same study, they also examine whether solar water heaters have a similar premium as PVs but could not statistically confirm this. Desmarais (2013) find that PVs increased the value of houses and decreased the marketing time in her 30 case studies in Denver. Using a sales comparison analysis on 23 properties in Oregon, Watkins (2011) also finds that there is a positive premium. In their study covering almost 23,000 houses in eight states in the USA during 2002-2013, Hoen et al. (2017) also confirm that houses with PVs are consistently traded with an average premium of $4 per installed watt.

Income and replacement cost methods have been suggested as alternative approaches when using the comparable sales method is not possible. However, studies using these methods are very limited. Klise et al. (2013) developed a PV value tool to address the value of the electricity-generating capabilities of a PV system and to overcome the lack of comparable market information. In addition to the PV value tool, Adomatis and Hoen (2016) compare the results of income (DCF) and replacement cost approaches with the paired sales method in 43 PV houses in six different states in the USA and the results from previous hedonic regression analyses. Based on a paired sales analysis, they find that PV systems garnered an average premium of $3.78 per installed watt and the premium was close to the PV cost estimate. The income approach resulted in more conservative values for PVs compared with paired sales and hedonic regression analysis; the authors conclude that cost and income approaches can be used if other approaches are not possible.

2.4 Defects of the reviewed literature

Most of the certificate studies have relied on statistical analyses based on large data sets of certified and noncertified commercial properties in the US markets relying mostly on EcoStar. Relying on the same data, these studies may suffer from systematically biased data. PV house studies also focus on the US market and may potentially have similar limitations as certificate studies. Studies on value increase through on-site energy production in commercial properties are virtually non-existent. Because professional property investors’ way of thinking and conducting value calculations differ greatly from that of house owners, studies focussing on commercial properties are needed. Furthermore, there are few studies that investigate on-site energy production other than PVs. Nor does certificate research separate the impact of on-site energy production from other green factors. The results of the existing studies are difficult to apply to individual properties because of the heterogeneity of markets and assets. Furthermore, in small markets, an insufficient amount of comparable transactions limits the usefulness of statistical analyses. Therefore, it is important to understand how sustainability affects cash flow parameters in the discounted cash flow framework and, consequently, the value of an individual property. The guidance note of the RICS (2013) also mentions that DCF calculation may reflect sustainability issues faster than other appraisal methods. DCF enables surveyors to reflect more closely on the attributes that do not impact the market value yet but that may have an influence in the future.

Vimpari and Junnila (2014) and Christersson et al. (2015) seem to be the only practitioners who have considered value-influencing mechanisms of green certificates and energy efficiency, respectively, in the discounted cash flow framework. All the respondents in Vimpari and Junnila’s (2014) study valued the case property with a certification more highly than without a certification. The main reasons for the observed average value increase of 9% in the certified property were improved net operating income and lower discount rate, which is in line with earlier research. Christersson et al. (2015) studied energy efficiency investments in almost 30 case properties, finding an average property value increase of 2.5%. Popescu et al. (2012) claim that in addition to energy savings, property value increases are essential in the assessment of the payback period in energy efficiency investments. However, the current study is the first one to elaborate on the value-influencing mechanism of on-site energy production from that of professional property investors’ perspective in the discounted cash flow framework. We also claim that the location of the property significantly affects the value of on-site energy production, which has mostly been ignored in previous (on-site) energy feasibility studies. On-site energy systems in more appealing locations have more value to investors than systems in more remote locations. The value generated through location can be incorporated into the analysis by using the property’s yield as a discount rate when capitalising energy savings into the value of the property (Leskinen et al., 2020; Vimpari and Junnila, 2019; Vimpari and Junnila, 2017; Vimpari and Junnila, 2017).

3.1 Description of the case study and methodology

A Finnish investor provided data for one of their logistic properties located in the Helsinki metropolitan area. The logistics centre was originally built in 2013 and has a leasable area of approximately 20,000 m². The property underwent an extensive energy renovation in 2018: district heating was replaced with a ground source heat pump system, and 480 kWp of rooftop PVs were installed. The ground source system covers 95-99% of the required energy, and the rest is covered with an on-site electric boiler that generates heat directly from electricity. Additionally, the system includes a 5 m³ hot water reservoir for balancing heat peak loads, i.e. the ground source system heats up the reservoir, which can then release energy for heat peak loads to avoid using the higher cost electric boiler. Non-energy generating investments, such as installed light emitting diode lights and other investments that improve the energy efficiency of the case property, were excluded from this examination. Historical performance data (Actual measurement data, 2019) for the first 10 months (November 2018-August 2019) of operations were provided together with the current pricing of heat (Local energy company, 2019c) and electricity (Local energy company, 2019b). To have a full year of performance data, the past two months (September-October 2019) were forecast based on the property’s historical consumption data. This data was used to calculate the annual savings for both heat and electricity with the following equations:

The impact of on-site energy production investments on property value was calculated using a 30-year discounted cash flow analysis. A 30-year cash flow was selected based on the typical lifecycles of PVs and ground source heat pumps (Branker et al., 2011). The discounted cash flow equation clearly shows that the value of the property increases through the capitalisation of the improved net cash flow because of a decrease in operating expenses, as the IVS (2017) suggests:

Accordingly, the increase in property value generated through savings in the operating expenses can be expressed as follows:

The present value of net savings over the 30-year lifecycle of the on-site energy investment was estimated based on the first year’s gross savings extracted by the maintenance costs and capital expenditure of the investment. Because of confidentiality reasons, we did not have access to the actual maintenance costs. In this paper, the estimated costs are based on the findings of earlier research:

• The maintenance cost of the PV system: 13.5 €/kWp (Vimpari and Junnila, 2019);

• The maintenance cost of the ground source heat pump: 5.4 €/kW (Kontu et al., 2020);

• The capital expenditure of replacing the inverter of the PV system in the year 2015: 0.05 €/kWp (Fraunhofer Institute, 2019); and

• The capital expenditure of the ground source heat pump in the year 2015: 30% of the initial investment cost (Kontu et al., 2020).

In the base scenario, electricity prices are assumed to increase by 4.1% p.a. and district heating prices 4.4% p.a. based on the historical price growth rates during 2000-2018 (Statistics Finland, 2019b). During the same period, the growth of the consumer price index has amounted to 1.5% (Statistics Finland, 2019a), which has been used as the inflation rate. The net savings were discounted with the sum of the prime logistics yield of 5.8% in the area (Datscha, 2018) and the inflation rate (7.3%). Four logistics tenants occupy the case property with gross lease agreements, meaning that the owner will directly and immediately benefit from the savings generated by the on-site energy production system.

3.2 Description of the survey to valuation professionals

According to the reviewed past research and IVS (2017), on-site energy production should be valued so that at least decreased operating expenses will be capitalised into the value of the property, as explained in 3.1. However, because valuing renewable on-site energy as a part of the property’s cash flow is a fairly new concept, both in research and practice, a survey targeted at valuation professionals in the Finnish real estate market was conducted as a part of the current study. The online survey was sent to authorised property valuers (altogether app. 270 in Finland) via the Finland chamber of commerce’s mailing list and to the members of the Finnish association for real estate valuation (250 members, of which half are authorised) in October 2019. The recipients hold several kinds of positions in surveying, consultancy and investment organisations. They were also requested to share the survey link with their colleagues working with property valuations to expand the number of potential respondents.

Altogether, 73 property valuation professionals responded, which is a good sample size considering the number of valuers in Finland. Approximately, 80% of the respondents worked as surveyors and the rest in investment management, consultancy and real estate development. The majority of the respondents were authorised property valuers (app. 85%), and almost 90% had more than five years of experience in property valuation. Experience in valuations with PVs and heat pumps was rare, with almost 60% having no experience with PVs and over 40% having no experience with heat pumps. The survey consisted of questions investigating the current valuation practice, namely, how on-site energy production affects the property’s cash flow parameters. The valuers were first asked to evaluate how on-site energy production affects the property’s value in the discounted cash flow context in general, specifically in the Finnish real estate market, and then in the presented case study (Figure 1).

4.1 Case study results

Figure 2 presents both the production numbers (MWh) and savings (€) of the on-site energy investment for the first operating year. The first year’s net savings on operating expenses of the on-site energy totals €101,000 of which 70% is because of the ground source heat pumps and 30% because of the of PVs. When estimated over the 30-year lifecycle of the system and capitalised with the prime logistics yield in the area [5.8% (Datscha, 2018)], the property value increases by €2.048m. The total capital expenditure for the system was €1.560m, meaning that the property owner immediately gains a benefit of approximately €490,000 from the investment. Compared with the results of traditional investment calculation methods, such as a payback time of over 14 years, the investment is very appealing from a real estate investment perspective. The return of the on-site energy investment amounts to 7.1%. The value increase of the property exceeds the investment costs when the return of the on-site energy investment is the same or exceeds the property’s yield.

In addition to the base scenario, where energy prices were assumed to grow at the same rate as during 2000-2018, a sensitivity analysis with high and low energy price scenarios was calculated. In the low energy price growth scenario, energy prices were assumed to grow at the same rate as inflation (1.5%). Then, the value increase of the case property would total €1.396m. In the high energy price growth scenario, energy prices were assumed to grow 2% points faster than in the base scenario. Then, the value increase of the case property would total €2.707m.

In addition to the savings in operating expenses, the sustainability of the case property increases. The on-site energy investment reduces the annual CO₂ emissions by approximately 430 t CO₂, based on average emissions of 247 kg CO₂/MWh for the district heating system (Local energy company, 2019a), 158 kg CO₂/MWh for electricity system (Motiva, 2019) and 35 kg CO₂/MWh for solar energy (Fthenakis et al., 2008).

4.2 Survey results

The survey respondents were asked to assess whether climate change affects property values. Approximately, 23% of the respondents were of the opinion that climate change has no impact on property values, even in the long run. This is surprising considering the lively and ongoing debate regarding climate change threats and the ever-tightening environmental regulations. The valuation professionals were also asked if renewable energy production is reported in valuations. Here, 51% of the respondents answered that renewable energy production is reported seldom or never as part of the valuation process. However, 90% of the respondents say that the existence of on-site energy production in properties already is or will become the norm; this indicates the need to develop reporting practices. Furthermore, it is an interesting question of how this change should be considered in all valuations (Figure 3).

The valuation professionals were asked what kind of impact on-site energy production has on different cash flow parameters in general in the Finnish market. The influence level was measured using a five-point scale ranging from “significant decrease” to “significant increase” for the value of the different parameters. Figure 4 shows that respondents felt that on-site energy production had the largest positive impact on the image of the property (84%). The pronounced importance of reputation is in line with the previous findings, for instance, how consumers make decisions that enhance their environmentally responsible behaviour ((Delmas and Lessem, 2014; Sexton and Sexton, 2014). The next largest positive impact was seen on decreased energy costs (78%), increased saleability (78%) and increased leasability (64%). Here, 46% saw that the risk of obsolescence should decrease and 32% said that the yield should decrease. Rents and rental growth instead were assessed as staying almost unchanged, with rent level even decreasing in the overall results (average), which is quite surprising. Because on-site energy production positively affects almost all cash flow parameters (rental growth being the only unchanged parameter), it is natural that the value of the property increases. Indeed, 71% of respondents agreed with the value increase, which is a smaller share than the answers related to operating expenses would indicate.

After a general assessment of the impacts of on-site energy on cash flow parameters, the survey presented the case study and asked the valuation professionals to evaluate whether the savings are fully capitalised into the value of the case property and if on-site energy production also affects other cash flow parameters. Almost, 80% of respondents stated that the savings in the operating expenses are fully capitalised into the value of the case property, corresponding well with the general assessment. The rest approached the question a bit more conservatively, considering only a part of the savings in the capital value of the case property. Only one respondent would not capitalise on the savings at all. These findings confirm the presented value-influencing mechanism. However, changes in other cash flow parameters were seen in slightly different ways in the case property than in the general assessment. Approximately, 65% of the respondents claimed that the market rent level will not change because of the investment, while 26% saw a slight increase. Three-quarters of the respondents expected rental growth to remain unchanged, and only 16% expected a slight increase. Both of these variables were impacted slightly less in the case study than in the general assessment. Perhaps, some of the respondents assessed only the current lease agreements, even though the (future) market rent level was the topic of the question. Current leases were announced to be gross leases. In a gross lease structure, the lessor gains the benefits of savings in operating expenses, while in net leases, this is reversed, and lessees may be willing to pay higher rent to reflect the lower expenses. Eventually, the owner will extract the same value regardless of the lease structure (Reichardt, 2014; Fuerst and McAllister, 2011). On the other hand, better future leasability should have been reflected in the rental growth rate. The impact on the yield was estimated to be a bit more positive in the case study than in the general assessment: 49% claimed on-site energy has no impact on yield, while 40% felt that the yield should slightly decrease. This is rational because a slight decrease in the yield can be more easily justified when there is poor availability of market data on other cash flow variables.

Figure 5 breaks down the reasons for the increased leasability and saleability of the property. Increased leasability was seen to be mainly because of the property’s improved image and attractiveness. Further, the number of potential tenants was assessed as growing and the risk of obsolescence as decreasing even more than in the general assessment. These improvements should also affect the occupancy ratio positively. Increased saleability was clear because of an increase in the number of potential buyers. Some of the respondents commented that this was also the main reason they felt that it was reasonable to decrease the yield.

Those valuers (n = 49), who had experience in valuations with either heat pumps or PVs, saw on-site energy slightly more positively both in the general assessment and in the case study than the rest of the respondents (n = 24). However, the differences are not statistically significant.

4.3 The implications of the survey results on the case study

In the presented case study, the value increase of the property is based on the capitalised savings in the operating expenses. The survey results imply that there may be positive effects on other cash flow parameters as well. Improved attractiveness and image of the property, as well as a higher number of potential tenants, would imply higher rents. A higher number of potential buyers and decreased risk of obsolescence would imply a lower yield. The base assumption is that the yield is 5.8% and that the gross rent level is 10 €/sqm/month (median operating expenses in the area being 2.65 €/sqm/month) before the on-site energy investment; these figures are hypothetical but are based on prime logistics market data in the area (Datscha, 2018; KTI, 2017). Approximately, 30% of the respondents answered that the yield would decrease. Over half of them estimated the yield within the range of 5.5-5.7%. That 25 % of the respondents who claimed that the rent level should increase estimated the (gross) rent levels to be from 10.5 to 12 €/sqm/month. Table 1 shows the impact of the higher rent level and lower yield on the property value in addition to the capitalised savings.