The Rebound Effect in Home Heating

A guide for policymakers and practitioners

Ray Galvin

• Publisher: CRC Press
• Year: 2015
• Print ISBN:
  9781138788350
• eBook ISBN:
  9781317448310
• Edition: 1
• Topic: Architecture
• Subtopic: Urban Planning & Landscaping

This is a definitive guide to the rebound effect in home heating – the increase in energy service use after a technological intervention aimed at reducing consumption. It sets out what the effect is, how it plays out in the home heating sector, what this implies for energy saving initiatives in this sector, and how it relates to rebound effects in other sectors. The book outlines how the concept of the rebound effect has been developed and the scope of research on it, both generally and particularly in the home heating sector. Within the context of energy and CO2 emissions policy, it summarises the empirical evidence, exploring its causes and the attempts that are being made to mitigate it. Various definitions of the rebound effect are considered, in particular the idea of the effect as an energy-efficiency 'elasticity'. The book shows how this definition can be rigorously applied to thermal retrofits, and to national consumption data, to give logically consistent rebound effect results that can be coherently compared with those of other sectors, and allow policy makers to have more confidence in the predictions about potential energy savings.

1
The Rebound Effect and Domestic Heating

1. Setting the scene

In the six months from October 1973 to March 1974 the price of oil quadrupled, from US$3 to US$12 per barrel. Political issues, together with declining oil production in the United States, led to concern about high and unstable oil prices continuing into the future (Bardi, 2009). As one of a number of measures to enhance the security of energy supply, OECD countries initiated a regulatory process for higher energy efficiency standards in buildings, household appliances and other sectors of the economy.

The economist Daniel Khazzoom investigated the effects of these energy efficiency improvements on the consumption of energy in household appliances. In theory, an increase in energy efficiency should cause a corresponding decrease in energy consumption: doubling the energy efficiency should halve the energy consumption. In a paper that has become one of the most frequently cited in the field of energy studies, Khazzoom (1980) reported that in fact, the decrease in energy consumption was consistently less than predicted. He suggested increases in energy efficiency were not being fully utilised to produce savings in energy consumption. Instead, only a portion of these, if any, was being utilised for this purpose. The remaining portion was being syphoned off to increase the consumption of energy ‘services’.

Energy ‘services’ are the benefits people get from consuming energy: hours of watching television; indoor warmth on a cold winter’s day; kilometres (km) travelled in a car; products rolling off an assembly line. Khazzoom suggested people were increasing their consumption of energy services, as it was now cheaper to do so, since their appliances were now more energy efficient. Higher energy efficiency meant the price per unit of energy services had fallen. Consumers were able to take more energy services than previously, yet still pay less overall.

Khazzoom warned that energy efficiency improvements could therefore ‘backfire’ and lead to an increase in energy consumption, particularly in industry. Higher energy efficiency reduced the price of factory goods, which increased sales, leading to higher profits, providing an incentive for manufacturers to install more plant and produce even more goods.

The same phenomenon was observed by economist Len Brookes, working independently from Khazzoom (Brookes, 1990). Both called the phenomenon ‘backfire’ and both were concerned that energy efficiency improvements, which were aimed to reduce energy consumption, could have the opposite effect.

In later papers, Khazzoom (1987; 1989) noted that this phenomenon had been observed in the nineteenth century by the English philosopher and economist William Stanley Jevons. Jevons (1865) drew attention to the paradox that the more efficiently Britain consumed coal, the more coal it consumed – an observation that became known as the ‘Jevons paradox’ (see discussion in Alcott, 2005). In nineteenth century Britain, increasing the efficiency of steam engines and their associated machinery brought down energy costs per unit of production, making goods cheaper, leading to more sales, which increased manufacturers’ returns, enabling them to invest in more machinery and afford yet more efficient machines, thus consuming more coal to produce more goods. Increasing energy efficiency therefore led to more energy consumption.

Two prominent researchers of the 1980s disagreed strongly with Khazzoom’s and Brookes’ ‘backfire’ theory. Amory Lovins (1988) maintained that these counterproductive effects of energy efficiency improvements would be ‘insignificantly small’, though it is not clear what empirical evidence was offered for this claim. Grubb (1990) repeated this assertion, but also pointed to an important distinction between different types of energy efficiency improvements. There is, he suggested, a natural, continuous improvement in energy efficiency which is driven by the scarcity of resources and the desire for greater well-being. This will certainly lead to higher energy consumption, as its goal is economic growth. However, energy efficiency improvements that are specifically designed to save energy will not necessarily lead to significantly higher demand, he maintained, as they are not part of this growth cycle. They are designed to enable consumers to enjoy the same level of benefits as before, but for a lower price, consuming less energy.

Both Lovins and Grubb agreed that this will lead to some shortfall in the expected energy savings, but maintained this would be insignificantly small and would not lead to backfire.

It was in the midst of this debate that the terms ‘rebound’ and ‘rebound effect’ emerged. There is no certainty as to when precisely these terms began to be used. Lovins spoke of ‘rebound’ in this context in a brief paper in 1988 (Lovins, 1988), and the term ‘rebound effect’ was used in a conference discussion as early as 1983. Berry and Hirst (1983) reported that a conference delegate had spoken of a ‘rebound effect’ caused by householders who ‘raise their thermostat settings after completing conservation retrofits’ on their homes (p. 78). This is a good example of one common cause of the rebound effect in domestic heating.

By the early 1990s the study of this phenomenon had entered mainstream energy and economics research, initially under the rubric of the ‘Khazzoom– Brookes postulate’, a label coined by Saunders (1992). During the 1990s the use of the terms ‘rebound effect’ and ‘backfire’ became standardised in academic literature. ‘Backfire’ came to be used for cases where an energy efficiency improvement led to an increase in energy consumption. The more subtle situation was called the ‘rebound effect’ – where there was a reduction in energy consumption, but not as much as would have been expected, given the size of the energy efficiency improvement.

A further point introduced by Khazzoom (1980, 1989) was that the rebound effect can reverberate throughout an entire economy. An increase of energy savings and energy services in one sphere, such as manufacturing, can lead to cheaper products and more spare cash, which can be spent on other energy consuming activities. This ‘economy-wide’ or ‘macroeconomic’ rebound effect has been extensively discussed and investigated in more recent years, since computing power became available to track rebound effects in detail throughout the major sectors of an economy (e.g. Barker et al., 2007), or indeed the world economy (Barker et al., 2009).

2. Domestic heating, energy efficiency, energy services

2.1 Energy efficiency

One of the key terms in understanding the rebound effect is ‘energy efficiency’. Energy efficiency is a measure of how effectively the consumption of fuel is turned into the benefits that human beings want. Some devices can be given a precise, absolute energy efficiency figure. For example, a modern condensing boiler typically achieves 90 per cent efficiency. This means that 90 per cent of the energy consumed in the fuel driving the boiler is successfully converted into heat to warm the building. Older, conventional boilers typically have efficiencies of around 70–80 per cent, and the efficiency of wood stoves has risen from 30 per cent to up to 90 per cent.

Other devices and systems cannot be given an absolute efficiency figure, but instead are given figures relative to each other. A house or apartment block is an example of this. A house acts as a composite system, with energy being brought in and consumed to heat it, and the heat then being lost to the outside world by conducting and radiating out through walls, floor, roof and windows, or drifting out through cracks and openings. Efficiency for a dwelling is therefore a more complex concept than for just a boiler or wood stove. The better a dwelling retains its heat, the more efficient it is. For example, a dwelling which requires 100 kilowatt-hours of energy consumption per square metre of floor area per year (kWh/m²a) to keep a steady, indoor temperature of 20°C is twice as efficient as a dwelling that requires 200kWh/m²a to keep the same indoor temperature.

Throughout this book, energy efficiency is almost always treated as a relative metric, rather than an absolute metric. Since the book will always be dealing with changes in energy efficiency relative to what they were before the change, this makes good mathematical sense and simplifies things greatly (see Appendix, Section A.1).

The efficiency of home heating has been steadily increasing over the last 40 years, particularly since the first oil shock in 1973–74, and even more sharply since concern about climate change became widespread late last century. A home built to the minimum standard of the building code in most western European countries in 2014 is about four times as efficient as one built to the minimum standard in the mid-1970s, when countries first included energy efficiency requirements in their building codes (Galvin, 2012). This means it is designed to produce the same level of thermal comfort for one-quarter of the energy consumption. In Germany for example, the building regulations in 1977 set the maximum permissible energy consumption for domestic heating at an average of 270kWh/m²a. This was reduced progressively over the following decades, until the most recent tightening of the regulations in 2009 brought it to 70kWh/m²a. The energy efficiency requirement for German homes today is therefore 3.86 times as high as it was (or 2.86 times higher than it was) in 1976.

2.2 Energy services

A further concept that is essential for understanding the rebound effect is that of ‘energy services’. Energy services are the benefits that people derive from consuming energy. In the home these include a large number of benefits which can be identified under the broad headings of thermal comfort, electronic entertainment, cooked food, cleaned clothes and dishes, and the operation of devices that enable people to work from home. These are not hard and fast divisions. Cooking is often part of entertainment, while computers can be used for both work and entertainment. The benefits relating to domestic heating come under the broad heading of thermal comfort. This includes air and radiant heating, evenness of warmth, air humidity level, air purity, heating up and cooling down time, and level of breeze or draught. Cooling is also an aspect of thermal comfort, but is a subject in itself and is not covered in this book.

Energy services (such as thermal comfort) are the reason energy is actually consumed by people. When planning a new building or an energy efficiency upgrade on an existing home, engineers need to know what level of energy services they are aiming to achieve. Governments therefore set certain standards of thermal comfort, which new buildings have to attain. Many of these are based on or draw upon the standards published by ASHRAE (American Society of Heating, Refrigerating and Air-Conditioning Engineers), which attempt to define the level of indoor temperature, light, humidity, etc., which is comfortable and healthy for the average person (see discussion in Humphreys and Hancock, 2007). The German government has its own set of standards for thermal comfort, published by the German Institute for Standards (DIN – Deutsche Institut für Normung). These stipulate an indoor temperature of at least 19°C in all rooms all year round, with three full exchanges of indoor air per hour. A building which achieves these standards may be said to be providing 100 per cent energy services. All new builds have to be designed to achieve these conditions (100 per cent energy services) in Germany’s climatic conditions, without consuming more than the maximum level of energy consumption set down in the regulations for a building of that size and shape.

Energy services cannot be measured directly, as energy consumption can. If a building is not providing this level of energy services – say the indoor temperature is only 18°C and there is only one full exchange of air per day – it is very difficult to say what percentage of the full requirement of energy services it is providing.

A simple rule of thumb is used to address this problem. First, an energy performance rating (EPR) is calculated. This is the level of heating energy consumption that is required for this building to provide 100 per cent energy services. For new buildings, this will already have been calculated at the design phase, so as to conform to the regulations outlined above. Second, the building’s actual heating energy consumption, averaged over the last two years and adjusted for any unusual climate conditions over those years, is noted. The average level of energy services is then defined as the actual consumption divided by the EPR. If, for example, the EPR was 100 kilowatt-hours per square metre of floor area per year (kWh/m²a) and the actual consumption was 80kWh/ m²a, the level of energy services is taken to be 0.8, or 80 per cent.

This is not a perfect measure because it does not account for possible variations in types of thermal comfort. Nor does it account for the possibility that many people adapt to different temperatures, ventilation rates, etc., (see discussion in Chappells and Shove, 2007). However, as Galvin (2014a) explains in more detail, it provides a very useful rule of thumb for putting a number to the level of thermal comfort, and therefore energy services, being provided by a heating system and enjoyed by building occupants.

2.3 Energy efficiency and energy services

To understand the concept of the rebound effect, both energy and energy services have to be taken into account. These are two very different parameters. Energy services are the benefits people enjoy. Energy is the commodity that is consumed to provide these benefits.

Things can become confusing because the word ‘consumption’ applies to both. People ‘consume’ energy services: they ‘use up’ heat because it comes out of the radiator and escapes through the walls and windows, and they ‘use up’ fresh air by making it stuffy so that it has to be replenished. They also consume energy: they ‘use up’ the energy in the gas, oil or electricity that powers their heating systems. It is the interplay between these two commodities – energy services and energy – plus the parameter of energy efficiency, that leads to the rebound effect.

It was mentioned in 2.1 that new homes in Germany today are just under four times as energy efficient as homes built in 1977. However, today’s new homes in Germany do not consume a mere one-quarter of their 1977 counterparts’ consumption. Recent research indicates they consume, on average, about half the 1977 level (Schröder et al., 2011; Sunikka-Blank and Galvin, 2012; Walberg et al., 2011). The main reason for this discrepancy...

Table of contents

• Cover
• Title
• Copyright
• CONTENTS
• List of figures and tables
• Preface
• Abbreviations and symbols
• 1 The rebound effect and domestic heating
• 2 What causes the rebound effect in home heating?
• 3 The prebound effect
• 4 Methods for estimating the rebound effect in domestic energy consumption
• 5 Rebound effects in low energy dwellings and passive houses
• 6 Fuel poverty and the rebound effect
• 7 Rebound effects in non-residential buildings
• 8 Conclusions, insights and recommendations
• References
• Appendix
• Index