Insight in actual energy use in houses as a basis for energy performance contracting

I

NSIGHT IN ACTUAL ENERGY USE IN HOUSES AS A BASIS FOR

E

NERGY

P

ERFORMANCE

C

ONTRACTING

Prof. dr. ir. Henk Visscher OTB research for the Built Environment,

Faculty of Architecture and the Built Environment, Delft University of Technology P.O. Box 5030, 2600 GA Delft

I

NSIGHT IN ACTUAL ENERGY USE IN HOUSES AS A BASIS FOR

E

NERGY

P

ERFORMANCE

C

ONTRACTING

ABSTRACT

Energy performance contracting is based on the idea that the investments for the improvements of the energy performance of a building through better insulation and more efficient installations will be covered by the savings of the use of energy in the use phase. The external contractor takes care of all the investments costs and guarantees the owner a fixed energy bill and hopes to profit from it. An interesting concept that is widely advocated as a solution to deal with some problems that mostly obstruct energy efficiency renovations. In the EPC approach the owner does not have to invest and does not take any risk. Still, at least in the Netherlands, it is not often used in practice and there are indeed some preconditions for projects that have to be met to make the approach work. The most important precondition is that the actual energy savings in practise will be enough to cover the investments with a reasonable payback time. This paper presents some results of research projects that compared the expected energy savings according to the energy performance calculation tools with the actual energy use in practice. The results of these projects show a large discrepancy between expected and real energy use caused by various reasons: design and construction failures and the influence of behaviour. It appeared that in the houses with a bad energy performance the behaviour led to far less energy use that expected and in the houses with a very good energy performance to a higher energy use because of rebound effects. What does this mean for energy performance contracting?

Keywords: Actual energy use; Energy efficiency; Houses; Energy performance

contracts.

1. I

Energy performance contracting is based on the idea that the investments for the improvements of the energy performance of a building through better insulation and more efficient installations will be covered by the savings of the use of energy in the use phase. The external contractor takes care of all the investments costs and guarantees the owner a fixed energy bill and hopes to profit from it. An interesting concept that is widely advocated as a solution to deal with some problems that mostly obstruct energy efficiency renovations. In the EPC approach the owner does not have to invest and does not take any risk. Still, at least in the Netherlands, it is not often used in practice and there are indeed some preconditions for projects that have to be met to make the approach work. The most important precondition is that the actual energy savings in practise will be enough to cover the investments with a reasonable payback time.

In Europe the residential sector is considered to be responsible for 30 per cent of the total energy consumption. The energy saving potential of the building stock is large and is considered to be the most cost efficient sector to contribute to the CO2 reduction ambitions. The Energy Performance of Buildings Directive (EPBD) is the driving force for all member states to develop and to strengthen energy performance regulations for new buildings and energy performance certificates for the existing stock (Mlecnik e.a. 2010; Murphy e.a. 2012). The goals are to build net zero energy buildings in 2020 and to reach a neutral energy situation in the whole stock by 2050.

  Various studies in the residential sector showed that the impact of the policies and regulations are often not as expected. Theoretical energy use that is calculated on base of the standards differs largely from the measured actual energy use. The approach and the findings of two research projects in the Netherlands will be presented in the next sections. The first deals with the effectiveness of energy performance regulations for new dwellings. The second evaluates the relation between the indicated theoretical energy consumption according to the issued energy performance certificates compared to actual energy use in the overall Dutch housing stock. In the last section these results are interpreted and the consequences for Energy performance contracts are discussed.

2. E

N

In the Netherlands, 1995 energy performance regulations were introduced in the national building regulations. It consists of a calculation method laid down in a national standard called EPN (energy performance norm) and a limit value, the energy performance coefficient. A lower co-efficient stands for a higher energy performing building meaning that the envelope and installations provide conditions that the same internal temperatures can be reached with less energy use. Since its introduction the EPC was sharpened several times. It started at 1.5 in 1995 and is now on the level of 0.6. In 2015 it will be lowered to 0.4 and in 2020 it should reach 0, indicating an nearly zero energy situation. The EPC is a non-dimensional digit. All building characteristics and installations that affect the energy demand for space and hot water heating, ventilation and lighting (of communal spaces) are incorporated in the calculation of the energy index (EI), which is the basis for the EPC. A further explanation of the calculation method can be found in Majcen (2013a).

Figure 1 Yearly gas consumption in m3 in Dutch dwellings (WoON 2009)

Since the introduction of energy performance regulations in 1995 only a few representative statistical studies were conducted to assess the effect of the regulations on the actual energy use. The samples were of limited size as well. In two of these studies no statistical correlation was found between the energy performance coefficient level and the actual energy use per dwelling or per square meter. In the analysis of the WoON (2009) survey, carried out on behalf of the Ministry for Housing, Planning and the Environment in 2006 and containing a sample of 5000 dwellings which is representative for the Dutch housing stock, also no correlation was found between the different levels of the energy performance coefficient and the actual energy use per dwelling and per square meter. Figure 1 shows the data on yearly gas consumption in

  Dutch dwellings in 2005 per construction year class. Remarkable is the increase in use per square meter for the newest houses that were built under the strictest regulations. In another research, Guerra Santin (2009, 2010) compared the actual and expected energy consumptions for 313 Dutch dwellings, built after 1996. A combined approache generated very detailed and accurate data of the (intended) physical quality of the dwellings and installations, about the actual energy use (from the energy bills) and of the households and their behaviour. In energy inefficient buildings with a high energy performance coefficient, actual energy consumption for heating was almost twice lower than expected, whereas in buildings with a low energy efficiency the expected and actual energy use coincided much better. She found that building characteristics were responsible for 19 to 23% of the variation in energy used in the recently built building stock. Household characteristics and occupant behaviour seemed to be responsible for 3 to 15% of the total variance. Neither our study nor the studies found in the literature allow to state that building characteristics, household characteristics and occupant behaviour altogether are responsible for more than 38% of the variation on energy consumption of dwellings built after 1995. Therefore at least 62% of the variation in energy use was unexplained by theoretical performance and behaviour and must have other reasons. There are indications that the explanation for this remaining part could be related to the fact that buildings practically are constructed differently than is described in official documents and to HVAC services operating in very different conditions than assumed. A report by Nieman (2007) showed that in a sample of 154 dwellings, 25% did not meet the energy performance requirements: the energy performance was incorrectly calculated. Nevertheless the building permit was issued. In 50% of the dwellings, the realization was not in accordance with the design. These results are supported by earlier findings about inadequate performance of building control in the Netherlands and other countries (Heijden, JJ van der, e.a. 2008, Meijer F. e.a. 2006, Meijer e.a. 2010, Visscher e.a. 2011). Gommans (2008) monitored for 17 years the energy performances of energy efficient buildings. 40% of solar boilers appeared to function poorly. Only 25% of the heat pumps reached the expected efficiency. This was essentially due to realization faults, lack of control and lack of continuous monitoring. Another study by Elkhuizen e.a. (2006) in office buildings showed that up to 28% of the energy could be saved by better quality assurance.

3. E

Although new buildings make it easier to apply energy saving measures, the largest energy saving potential is in the existing building stock. On average new dwellings add less than 1 per cent per year to the housing stock. National and local governments have formulated ambitions, programmes policies and instruments to stimulate the improvement of the energy performance of the existing stock. The most important policy tool required by the EPBD in the European member states is the issuing of Energy Performance Certificates. All member states have to produce an energy label for a building at the moment it is sold or re-rented. This is not yet current practice everywhere, mostly due to lacking enforcement. In the Netherlands however, the whole social housing stock is labelled. The label indicates the energy demand for heating and cooling. The present label data base covers a large share of the housing stock in the Netherlands. This forms a basis to monitor the progress of the renovation practices. Besides that it is also useful to study the relation of the energy labels with the actual energy use. The progress of renovations and energy upgrading measures stays far behind expectations and formulated ambitions in 2008 when most of the policies, covenants and improvement programmes were set up. The social sector in the Netherlands is still relatively large (35%), well organised and relatively rich. A few years ago the sector formulated ambitious programmes, but these are nowadays scaled down because of several reasons. The economic crises reduced the financial position

  of the housing associations. The housing market also dramatically slowed down which also affected the funding for renovations because this largely depends on the sales of property. Also it proved to be difficult to get approval of tenants for renovations that require an increase of the rents (70% of the tenants have to agree). It is hard to assure the saving of energy costs resulting of the improvement of the dwellings.

Figure 2 Actual and theoretical gas consumption in Dutch dwellings (Majcen et al., 2013a) The actual energy use is largely influenced by the use and behaviour of the tenants. Some preliminary figures demonstrate the difficulty in ‘forcing’ reduced energy use by improvements of dwellings. The dwellings with the worst energy label (G) in practise use far less energy as expected, while the most advanced dwellings (A) use much more, probably due to a combination of the rebound effect and an increase in comfort level of the dwellings (Majcen et al 2013a 2013b) and underperformance of the buildings and installations. Figure 2 shows the actual and theoretical gas consumption per dwelling per energy label.

In the home owner sector the issuing of energy labels stays yet far behind. Although they were mandatory, until now there has not been an enforcement system. Energy labels will become common practice and affect the sales price. Still there are no obligations foreseen to make improvements and higher labels mandatory. It is hard to require investments and property rights are probably an obstruction. Still there are some ideas for taxation measures. Bad labels could be punished with higher transaction taxes or higher property taxes than good labels.

4. M

Tigchelaar et al. (2011), calculated a ‘heating factor’ (actual heat demand divided by theoretical demand). The average heat factor in a sample of 4700 representative dwellings was found to be below one, meaning that the expected energy use is overestimated. Cayre et al. (2011) studied expected and actual energy use in 923 French dwellings and reached similar conclusions. A similar result was also discovered by Hens (2010), who monitored the actual use of two types of dwellings in Belgium (from the 80s and 90s). He found that the consumption on average was only half of the

  calculated energy use. On the other hand, in 12 multi-family buildings in Austria that were renovated, Haas (2000) found that the actual energy consumption was significantly higher than the expected energy consumption. Similar results were obtained by Branco (2004) in a multi-family complex in Switzerland and in a similar sample in France (Marchio 1989). Based on these results, it seems that the theoretical energy use often is overestimated for average and less energy efficient dwellings and underestimated for new or retrofitted buildings. This phenomenon can partly be explained by the so-called rebound effect. Sorrell (2009) provides an overview of methods for calculating the rebound effect and a summary of available studies. He concludes accordingly, in OECD countries, that the mean value of the long-run direct rebound effect is likely to be less than 30%. This means that up to 30% of the efficiency gained through technical improvements of building and appliances are turned into increased consumption (higher comfort) following from direct change in user behaviour.

5. C

E

P

C

The previous sections have shown the factors that can influence the actual energy use apart form the physical improvement of the building and the installations. If this is negelegted Energy Performance Contracting becomes a risky business and cannot be succesfull. Renovated houses show mostly a higher (up to 20%) energy use for heating and cooling than expected, caused by: rebound effects (higher average temperature settings) and under performance of the buildings because of design and constructions faults. Other findings that the behavior and comfort levels varies largely between different levels of energy performances of dwellings and that this differs very much (up to 50% less energy use) from what is expected according to energy performance standards. This means also that the average temperature over the whole dwelling is much lower than in worse insulated dwellings than we would expect. The consequences of this knowledge is that when a house is renovated the occupants enjoy a higher level of comfort, but that the actual energy savings are very moderate.

Figure 3 Factors to incorporate in Energy Performance Contracts

Figure 3 shows the factors that have to be taken into account in Energy Performance Contracts. If we would only look to the building performance label for a renovation form a G label to an A label, we would calculate a potential saving of 60 – 70 % (the blue

  line). If that would define the investment potential for the renovation then a fair payback time will be in reach. Taking the actual energy use as a stating point: the investment space would be very low (the pink line). So we will have to find a way of dealing with these factors. First of all one would have to deal with the quality assurance of the renovated buildings to make sure that the final quality is like it is required and is designed. It is more difficult to deal with the rebound effect and the increased comfort levels. The challenge is to find a way to separate the building related energy reduction form the energy used for an increased comfort (higher temperature levels for heating, or lower in case of cooling). The Energy Performance Contract with the owner of a building should deal with the improved performance of the building and installations and separately the user of the building should remain to pay for an increased comfort (difference between the pink and the yellow line).

6. R

Branco, G., Lachal, B., Gallinelli, P., Weber, W. 2004, Predicted versus observed heat consumption of a low energy multifamily complex in Switzerland based on long-term experimental data, Energy and Buildings, 36:6, 543-555.

Cayre, E., B. Allibe, M. H. Laurent, D. Osso 2011, There are people in this house! How the results of purely technical analysis of residential energy consumption are misleading for energy policies, Proceedings of the European Council for an Energy Efficient Economy (eceee) Summer School, 6–11 June 2011, Belambra Presqu'île de Giens, France.

Elkhuizen P., Scholten J.E. and Rooiakkers E. 2006, Quality Control of HVAC services: evaluation of existing instruments and a vision for the future, TNO bouw/Halmos report for Senter Novem.

Gommans L.J. 2008, Energy performances of energy efficient buildings, TVVL magazine, September 2008, pp. 18-24.

Guerra Santin O., Itard L. and Visscher H.J. 2009, The Effect of Occupancy and Building Characteristics on Energy Use for Space and Water Heating in Dutch Residential Stock, Energy and Buildings, 41, 1223-1232.

Guerra Santin O. and Itard L. 2010, Occupant behaviour in residential buildings in the Netherlands: Determinants and effects on energy consumption for heating, Building Research & Information, 38:3, 318-338

Haas R., P. Biermayr 2000, The rebound effect for space heating Empirical evidence from Austria, Energy Policy, 28:6, 403-410.

Heijden, J.J. van der, Visscher, H.J. & Meijer, F.M. 2008. Problems in enforcing Dutch building regulations. Structural survey, 25:3/4, 319-329

Hens, H., Parijs, W. and Deurinck, M. 2010, Energy consumption for heating and rebound effects, Energy and Buildings, 42:1, 105-110

Majcen, D., Itard, I. and Visscher, H.J. 2013a, Theoretical vs. actual energy consumption of labelled dwellings in the Netherlands: Discrepancies and policy implications, Energy Policy 54, 125 – 136.

Majcen, D., Itard, L., Visscher, H.J. 2013b, Actual and theoretical gas consumption in Dutch dwellings: What causes the differences? Energy Policy, 61, 460-471.

Marchio, D., A. Rabl 1991, Energy –efficient gas heated houing in France: Predicted and observed performance, Energy and Buildings, 17, Pages 131 – 139.

Mlecnik, E, Visscher, H.J. & Hal, J.D.M. van 2010, Barriers and opportunities for labels for highly energy-efficient houses. Energy policy, 38:8, 4592-4603.

Meijer, F.M., Visscher, H.J., 2006, Deregulation and privatisation of European building-control systems, Environment and Planning B: Planning and Design, 33:4, 491–501. Meijer, F.M., Visscher, H.J. & Costa Branco De Oliveira Pedro, J.A. 2010 Building control systems of European Union countries: a comparison of tasks and responsibilities. International journal of law in the built environment, 2:1, 45-60.

  Murphy, L.C., Meijer, F.M. & Visscher, H..J 2012. A qualitative evaluation of policy instruments used to improve energy performance of existing private dwellings in the Netherlands. Energy Policy, 45, 459-468.

Nieman 2007, Final report Housing quality indoor environment in new built dwellings, Vrom inspectie Regio Oost, Arnhem.

Sorrell, S., Dimitropoulos J., Sommerville M. 2009, Empirical estimates of the direct rebound effect: A review, Energy Policy, 37:4, 1356-1371

Tigchelaar, C., Daniëls, B., Maenkveld, M., Obligations in the existing housing stock: Who pays the bill?, Proceedings of the European Council for an Energy Efficient Economy (eceee) Summer School, 6–11 June 2011,Belambra Presqu'île de Giens, France.

Visscher, H.J. & Meijer, F.M. 2011, Energy saving goals require reform of building regulations and control. In les Ruddock & P Chynoweth (Eds.), COBRA 2011 Programme and Abstracts RICS Construction and Property Conference 328-335. Manchester, University of Salford.