A drafty home has more risk than just catching a cold

Rabobank offers new and existing homeowners with green-labeled homes a discount on their mortgage interest to encourage energy efficiency and sustainable housing. An interest-rate discount is part of the bank’s ambition to contribute to the huge sustainability undertaking facing the housing market. However, interest-rate discounts are also given because of the assumed difference in risk between homes with red and green energy labels. The energy label does affect the residual debt risk: the risk that in case of a sale, the value of the home will be less than the remaining mortgage debt. This is because the energy label affects the value of a home. It was also assumed that households in green-labeled homes would be less likely to be in arrears on their mortgage payments. After all, their monthly energy bills are lower and they are less exposed to price increases. This assumption has however not been researched to any great extent. A new study by researchers at Tilburg University in association with Rabobank and based on data from the NVM and Rabobank shows that this is not really the case. Owners of green-labeled homes are not less likely to be in arrears on their mortgage payments than those in homes with no definitive energy label. Owners of red-labeled homes are slightly more likely to be in arrears on their payments than those in homes with no definitive energy label. This effect is however very low. It is therefore the difference in the risk of a larger residual debt - as well as the aspiration to contribute to making the housing stock more sustainable - that can justify giving a discount (or surcharge) on the mortgage interest according to a home’s energy label.

There has been a big increase in the availability of ‘green’ loans and ‘green’ mortgages in recent years. Interest discounts are available for sustainable homes, but the conditions and the discounts vary from one mortgage provider to another. An interest discount can encourage those looking for a home to choose an energy-efficient, green-labeled home over a red-labeled home and also encourage homeowners to improve the sustainability of their existing homes.

In this Economic Report, we focus on the relationship between energy labels and the mortgage risks for homeowners and mortgage providers arising from the arrangement of a mortgage. There are two main risks. Firstly, there is the risk that a homeowner will be unable to pay or continue to pay their monthly mortgage payment. Secondly, there is the risk of a residual debt, with the amount of the residual debt partly depending on the value of the home. It is well known that a home’s energy label affects the value of the home and how this value will develop, and therefore also affects the residual debt risk. But little is known is known about the effect of the energy label on the risk that a household will not be able to keep up with its monthly mortgage payments. This Economic Report presents the results of a large-scale study of the effect of energy labels on the risk of payment arrears in the Netherlands, conducted by researchers at Tilburg University (Bohorquez Correa et al., forthcoming). And we argue that a red-labeled home involves more risks than simply catching a cold.

All residential properties in the Netherlands have had a provisional or definitive energy label since 2015. The energy label provides information on the energy efficiency of the home. For homes with no definitive label, an estimate of the home’s energy performance is made on the basis of features such as the year of construction and the size. Homes with A or B energy labels are considered to be relatively energy-efficient, and are also referred to as green-labeled homes in this article. Homes with F or G labels are red-labeled; they are not energy-efficient and therefore also require a lot of work to make them sustainable.

By providing homeowners and potential home buyers information on a home’s energy label, the government aims to encourage homeowners to invest in energy-saving measures. This is needed to achieve the aspiration to improve the sustainability of the housing stock. To realize the goals in the Climate Agreement reached in 2019, nearly 1.5 million existing homes will have to cease using gas by 2030. Owner-occupied homes must also on average have an A energy label by 2030, according to the Energy Agreement.

There is still much work to do. In 2018, only 46 percent, slightly less than half of all residential properties, had a definitive energy label according to the official registration system for building energy labels. The Housing Survey Netherlands (WoonOnderzoek Nederland, or ‘WoON’), a large survey carried out once every three years by the Ministry of the Interior and Kingdom Relations (BZK), showed that only 14 percent of all homes had a definitive energy label of A or higher. Although housing associations have been called the drivers behind improving the sustainability of the housing stock, more owner-occupied homes have an A label than rental homes (figure 1). Around 35 percent of all owner-occupied homes had a ‘green’ label (B or higher) in 2018. So there is a long way to go before achieving the stated goal.

An energy label has implications for the potential mortgage risk (if a home is mortgaged) as well as a home’s sustainability. Arranging and providing a mortgage involves two main risks for the person taking the mortgage (the homeowner) and the mortgage provider. Firstly, there is the risk that the person taking the mortgage will not be able to continue making their monthly payments. Secondly, there is the risk that there will be a residual debt if the home is sold. These two risks can be linked: a homeowner in arrears on their payments may be left with a residual debt in case of a forced sale. Both these risks may be affected by the home’s energy label, in ways that we explore further below.

Energy labels and payment arrears

The energy label gives information on the monthly energy bill (table 1) and thus also perhaps regarding the risk of payment arrears. Owners of A label homes spent an average of € 168 per month on gas and electricity in 2018. This is € 30 less than the monthly energy costs for those living in a G-label home.[1]

But a high energy bill is not necessarily unaffordable for a household. One possible indicator of the affordability of the energy bill is the proportion of a household’s income that is spent on gas and electricity, or the energy costs to income ratio. According to the WoON, homeowners spent on average 6 percent of their income on energy in 2018.[2] Homeowners with a green (A or B) label on average spent a lower proportion of their income on energy than those living in red-labeled homes (F or G labels, table 1). It is notable that households living in a G-labeled home on average have higher mortgage payments (also in proportion to their income), than those living in A-labeled homes.

Households with a high energy costs to income ratio not only have relatively high outgoings, they are also more vulnerable to increasing energy prices. They feel the difference in their spending power more quickly. Energy bills have risen significantly since 2018. According to Statistics Netherlands (CBS), households are spending around € 334 more on energy this year than a year ago, with the average annual energy bill increasing by around 19 percent. This is partly due to the higher tax on energy. This kind of increase will be felt more by households in red-labeled houses with relatively high monthly energy bills.

Since owners of red-labeled homes spend a relatively large amount on energy (and on their living expenses), it is quite feasible that they will be more likely to have difficulty in making their monthly mortgage payments than those living in green-labeled homes. US Researchers Kaza and colleagues showed in 2013 that homes with an Energy Star (relatively energy-efficient homes) were indeed less likely to default on their payments The findings of this American study cannot however be translated directly to the housing market in the Netherlands. Arrears in mortgage payments occur less frequently in the Netherlands, due to the liability assumed when arranging a mortgage. This means that homeowners are liable for any residual debt, also in the event of a forced sale.

[1] In theory, this could be because homes with a G label are on average larger than homes with an A label. If we restrict the analysis to homeowners in single-family homes of between 120 to 150 m2 however, the differences remain. Households in an A-label home spend an average of € 170 per month on energy, compared to an average of € 196 per month for those in G-label homes.

[2] In accordance with Middelkoop et al. (2018), households with an energy costs to income ratio of less than zero and a negative disposable household income are excluded from the analysis. Energy costs to income ratios with values higher than 99.7 percent have been rounded to 99.7 percent. In line with the principles used by the CPB in its purchasing power calculations, households with very low incomes (less than 63 percent of the net minimum wage) are not included in the selection of respondents.

Is there also a link between the energy label and payment arrears in the Netherlands?

In order to test whether the likelihood of mortgage payment arrears is affected by the energy label, we used a unique longitudinal micro-data file with data on 73,782 Rabobank mortgages which were tracked between January 2015 and June 2018. These data were then supplemented with information from the Dutch Real Estate Brokers’ Association (Nederlandse Vereniging van Makelaars, or ‘NVM’) on energy labels and property features. Based on these data, it appears that there is no significant difference between homes with a ‘red’ (F or G) energy label and homes with a ‘green’ (A or B) energy label, an average energy label (C, D or E) or no energy label at all. The average house price, the amount of the mortgage, the household income and the LTV do not change in line with the energy label.

Probit regressions were conducted to estimate the probabilities of payment arrears. The dependent variable is a binary variable that can have two values: no payment arrears (zero, the reference category) or payment arrears (1). This analysis was conducted in the context of the doctoral research of Santiago Bohorquez Correa and his thesis supervisor Professor Dirk Brounen of Tilburg University.

The effect in the Netherlands appears to be low

It appears that there is a link between the energy label and the likelihood of payment arrears in the Netherlands as well. This link is through the energy costs to income ratio, but surprisingly applies only to homes with a red label. Adjusted for relevant mortgage and property features and income and demographic background characteristics, red-labeled homes are more likely to be in arrears on mortgage payments as the energy costs to income ratio increases. Compared to homes without a label, the likelihood of payment arrears for homes with a red label is approximately 0.026 of a percentage point higher with a very low energy costs to income ratio (a ratio of nearly zero, see figure 2). This effect increases to around 0.052 of a percentage point with an energy costs to income ratio of 20 percent. The effect is small, but significant. For green-labeled homes and homes with an average energy label, we find no significant effect of the energy label on the likelihood of payment arrears.[3]

[3] The effects of the energy label on the likelihood of payment arrears should be interpreted in comparison to the reference category: homes with no definitive energy label. On average, these homes can be compared to homes with an energy label D with regard to the year of construction, size and type, as a result of which the regression coefficients can be interpreted as the effects compared to an average home. Various regression models have been estimated; the choice of a different reference category had no significant effect on the conclusion that the energy label has little effect on the likelihood of payment arrears.

The fact that the energy label has only a minor effect on the likelihood of payment arrears could be due to various reasons. Firstly, we know that households adjust their energy use to the features of their home. Households in red-labeled homes are more likely to set their thermostats lower than those in green-labeled homes. And this means for instance that households in red-labeled homes use less gas (and therefore have lower energy costs) than would be expected on the basis of theoretical calculations, as shown by a study by PBL. Secondly, the study was conducted in a period in which the economy was strengthening and unemployment was falling. Together with the already high propensity of Dutch people to repay their mortgages, there were few homeowners in arrears on their mortgage payments during the study period. At the time of the study, there were only 1.3 percent of all mortgages with payment arrears at any one time.

Energy labels and the value of collateral

The second mortgage risk is what is known as residual debt risk. It is reasonable to assume that there is a link between the energy label and the amount of a residual debt, since the value of a home is to some extent determined by its energy label.

If the value of the home is lower than the mortgage debt on the home, this means there is potentially a residual debt. In this situation, the Loan-to-Value ratio is higher than 100 percent. A potential residual debt only becomes an issue if people move house, for instance due to divorce or separation or problems with meeting the monthly mortgage payments, for example due to losing one’s job. In a situation of payments in arrears, a mortgage provider will ultimately insist on a forced sale of the home. The ultimate sale price of the home will determine the amount of the residual debt.

It is reasonable to assume that households in green-labeled homes are less at risk of a sizable residual debt than households in red-labeled homes. The value of the home (and therefore also the sale price) is of course affected by the sustainability of the home. Based on Dutch data, homes with a green energy label sell faster and at a higher price than homes with a red label. This effect is seen mainly in areas outside the large cities, where the housing market features less demand. But the Netherlands is not unique. In other countries as well – including England, the United States and Spain – a positive connection has been demonstrated between the energy efficiency of a home and its value. Owners of red-labeled homes not only have to take account of a ‘brown discount’ (meaning that the home is worth less), but also less price appreciation, as shown by a study carried out in England. There is thus a possibility that red-labeled homes will lose value faster in the future.

Since a home’s value (and therefore also the potential proceeds of a sale) is to some extent determined by its energy label, the energy label does affect the size of any residual debt. A red energy label thus negatively affects the sale proceeds and can increase the amount of the residual debt.

A home’s energy label therefore tells us something about the potential mortgage risk, and in particular about the risk of a sizable residual debt, as well its energy performance. This is because the energy label affects the value of a home and therefore also the size of any residual debt: the mortgage debt that remains after the home is sold. The energy label has less effect on the risk of payment arrears. Owners of homes with a red energy label are however slightly more likely to get into arrears on their mortgage payments because they are more exposed to rising energy prices. But this effect is extremely small. Contrary to an American study, Dutch owners of green-labeled homes are not less likely to get into arrears on their mortgage payments than owners of homes with no label. This may to some extent be due to the fact that Dutch homeowners are among the most keen to repay their mortgages; by international measures, payment arrears in the Netherlands are much less common.

Large banks like Rabobank offer new and existing owners of green-labeled homes a discount on their mortgage interest. They do this partly because they want to contribute to improving the sustainability of the housing market, and partly due to the perceived differences in risk between red and green-labeled homes. The study described here is the first to demonstrate that in the Netherlands, there is little connection between the energy label and the risk of payment arrears. Owners of red-labeled homes are only slightly more likely to be in arrears with their payments. Due mainly to the difference in the size of any residual debt, green-labeled homes feature a lower risk for both those taking a mortgage and mortgage providers than red-labeled homes. From the perspective of variations in mortgage risk therefore, it is mainly the difference in residual debt risk that justifies a mortgage interest discount (or surcharge) according to a home’s energy label. The sustainability contribution is another reason for linking the rate of mortgage interest to a home’s energy label.

Ayala, A. de, Galarraga, I. & Spadaro, J.V. (2016). The price of energy efficiency in the Spanish housing market, Energy Policy, 94, 16-24.

Bohorquez Correa, S., Brounen, D. & Kok, N. (to be published). Energy efficiency as a predictor of residential mortgage risk, part of ongoing doctoral research at Tilburg University.

Brounen, D. (2016). De groeiende rol van energie-efficiëntie in vastgoed-financiering, chapter 7 in Berkhout, T.M., & Bartels, S. (Eds). Vastgoedfinanciering: klaar voor de toekomst. Amsterdam: Stichting Fundatie Bachiene.

Brounen, D. (2018). Ongunstig energy label doet weinig pijn in grote steden, newsletter from Tilburg University, 7 May 2018. Received on 16 July 2018.

CBS (2019) Energierekening 334 euro hoger, newsletter 16 February 2019.

Fuerst, F., McAllister, P., Nanda, A. & Wyatt, P. (2015). Does energy efficiency matter to home-buyers? An investigation of EPC ratings and transaction prices in England, Energy Economics, 48, 145-156

Fuerst, F., McAllister, P., Nanda, A. & Wyatt, P. (2016). Energy performance ratings and house prices in Wales: An empirical study, Energy Policy, 92, 20-33.

HIER verwarmt (2019). ‘Waar gebruik ik nu gas voor’, www.hierverwarmt.nl.

Kahn, M.E. & Kok, N. (2014). The capitalization of green labels in the California housing market, Regional Science and Urban Economics, 47, 25-34.

Kaza, N., Quercia, R.G. & Chao Yue, T. (2014). Home Energy Efficiency and Mortgage Risks. Cityscape, 16(1).

Koot, P., Vlekke, M., Berkhout, E., & Euwals, R. (2016). MIMOSI:Microsimulatiemodel voor belastingen, sociale zekerheid, loonkosten en koopkracht. The Hague: The Netherlands Bureau for Economic Policy Analysis (2011).

Long, J.S. & Freese, J. (2001) Regression models for categorical dependent variables using STATA. Section 4 Texas: Stata Corporation.

Middelkoop, M. van (2014). Energiebesparing: voor wie loont dat?Onderzoek naar de betaalbaarheid van energie en energiebesparing voor households. The Hague: Planbureau voor de Leefomgeving.

Middelkoop, M. van, S. van Polen, R. Holtkamp & F. Bonnerman (2018). Meten met twee maten.Een studie naar de betaalbaarheid van de energierekening van huishoudens. The Hague: Planbureau voor de Leefomgeving.

Rijksoverheid (2019). Klimaatakkoord hoofdstuk Gebouwde Omgeving.

Sociaal-Economische Raad (SER) (2013). Energieakkoord voor duurzame groei.

Stanga, I., Vlahu, R. & Haan, J. de (2017). Mortgage arrears, regulation and institutions:Cross-country evidence, DNB Working Paper no 580. Amsterdam: De Nederlandsche Bank NV (DNB).

Wijngaard, R. van den, Folkert, R. & Middelkoop, M. van (2014). Op weg naar een klimaatneutrale woningvoorraad in 2050.Investeringsopties voor een kosteneffectieve energievoorziening. The Hague: Planbureau voor de Leefomgeving (PBL) (2015).

Zancanella, P., Bertoldi, P. & Boza-Kiss, B (2018). Energy efficiency, the value of buildings and the payment default risk. JRC Science for Policy Report. Luxembourg: European Union

This version has been abridged on November 13, 2019