Energy Efficiency Policies around the World: Review and Evaluation

3.8 Energy Efficiency Obligations in Europe

Energy efficiency obligations is a recent and innovative measure whereby energy companies (supplier/retailer or distributor) have a legal obligation to promote and stimulate investment, which will save energy in customers' premises or households. When this obligation can be met by buying or selling the energy saving credits, it is usually called White Certificates.

Energy efficiency obligations are in some aspects similar to the older Demand side management (DSM) activities since there is an obligation on energy companies to undertake energy efficiency activities with their customers.  However compared to many of the 1990s DSM programmes, the obligations approach focuses on outcomes (i.e. energy saving targets) rather than money spent and has developed much cheaper monitoring and verification systems.

3.8.1 Existing Energy Efficiency Obligations in Europe

Six countries currently have energy efficiency obligations on energy companies in Europe: Belgium (Flanders Region), France, Italy, UK, Ireland and Denmark. Energy efficiency obligations are placed on energy suppliers in the Netherlands in the second half of 2007 and similar activities are under development in Poland and Portugal.

The approach to energy efficiency obligations on energy companies has developed differently with different mandatory parts of the energy industry and a wide variation in the end use sectors to which the obligations are applied (Table 3.4).  Nevertheless such obligations have been shown to be extremely flexible and have proved capable to work either with a traditional monopoly energy utility or in a fully liberalised market.
Table 3.5 provides details of the targets, their size and other key parameters of the energy efficiency obligations in four countries.

This review of energy efficiency obligations in Europe concentrates on the four major experiences which have been accumulative for some time and which are of a significant scale: Belgium, France, Italy and UK .

3.8.2 Experience from the Four Countries

Although there are many differences between the way the targets are set, the size of the targets, the concerned parties and the energy using sectors covered, there are in fact many similarities.  The common experience to date in the four European countries is analysed below using a common format .  However, the programmes do vary considerably in their nature, in the length of time that they have been running and the extent to which they have had independent evaluation, which is publicly available.  Therefore, it is not possible to cover all aspects to the same extent.

The size of the target, the end using sectors to which it applies, etc vary from country to country (Tables 3.4 and 3.5). However, the key principles are an obligation placed on an energy company by Government (or a Regulator) and a formal monitoring and verification process is enacted to ensure the targets are met by eligible energy saving measures.  Most countries have penalties for those companies which do not fulfil their energy efficiency obligations (Table 3.5).  In practice, no penalties have been issued as virtually all the energy companies have met their targets.

Target Sector & Size

Usually the size of the target and the sectors to be covered are decided by government rather than the regulator for that particular energy industry, although in some cases the Regulator is the appointed body to oversee and verify the energy efficiency obligations. Having national governments decide on the size of the obligation seems appropriate as energy efficiency obligations are linked to environmental concerns and have an important social dimension.  It is not easy for an unelected regulator to make judgements which are not solely based on economic grounds and can have a significant impact on energy bills in the short term.  The targets are set in relation to the volume of electricity or gas supplied or distributed.  In the residential sector, the simple proxy of customer numbers is often used rather than volumes of electricity.

The cost or implied cost varies considerably but even in the UK it is currently only ~1.5% of household fuel bills (Table 3.5,).

Most of the energy efficiency obligations allow "banking", e.g. carrying forward of excess savings from one target period to the next.  This has important benefits, not just for the energy company but also for the energy efficiency industries . 

In Flanders and the UK, the energy companies are required to ensure that there are also savings in low income households.  This is achieved by ring fencing a fraction of the energy saving target that has to be met by savings in such households.  In fully liberalised markets where the supply price controls to residential customers have been dropped, it is up to energy suppliers to decide how to cover the costs of the energy efficiency obligations.  Where prices are still controlled the Government could influence the way the tariffs recovered the costs and social tariffs could be exempted to address equity issues.

Interaction with Other Policy Mechanisms

Inevitably national governments have a variety of policies designed to improve energy efficiency in all end use sectors.  There can be complications from interactions between such policies, which either are required by legislation or are subsidised by central Government and the obligations on the energy companies.

In other words, is there either genuine collateral (in the case of existing legislation requiring Minimum Performance Standards of energy efficiency) or double subsidies of the measure by Government and the energy company?

A pragmatic approach has been taken to dealing with these issues in all countries.  For example, in the UK only energy efficiency measures, which produce a performance above, that required by legislation (e.g. in new build or major refurbishment or EU Minimum Performance Standards for Appliances) are approved as eligible energy savings and only for that part which is in excess of the regulatory requirement.  Indeed, the UK has gone further in the appliance field by only allowing an energy saving for an energy efficient appliance or heating boiler, which is above the market average for such products.

In a similar vein, the UK disallows the parts of the savings which are supported by any central government funding (e.g. programmes tackling low income households) .

In Italy and France, certain energy efficiency measures can be offset against income tax, but this has been allowed only as central Government funding/subsidies are not considered additional.

Nature of Target

There is widespread variation in the nature of the target set (Table 3.5).  There are local reasons for why this might be the case, and it is worth examining some of the key considerations.

The energy saving credits are based (or in a few cases measured) on the saving of delivered energy.  For countries concerned with reducing their energy imports, the use of primary energy which is usually 2.5 times delivered or final energy for electricity with the other fossil fuels being taken as equivalent to their delivered energy, is common.  For countries, which are concerned about reducing CO2, it is quite common for the energy savings to be carbon weighted by the carbon content of the saved fuel.

In terms of for how long the energy savings should be defined, the two extremes are simply an annual energy saving and lifetime energy savings Another issue for debate is whether the energy savings should be discounted over time to reflect the time value of money, as is common in normal financial appraisals.  The discount rates have varied between 8% and currently 3.5- 4%.  The key question is perhaps whether this is being done for economic or environmental reasons.  If for economic reasons, then the use of discount rates merely conforms to standard energy appraisal options.  However, if this is being done for climate change reasons, it may not be appropriate to discount the energy and consequently carbon dioxide savings, certainly not at a high discount rate . 

Definition of Eligible Measures

Again the eligible measures are usually defined in advance by the monitoring and verifying authority.   This means that only measures that have been independently proven to save energy are utilised.  In one sense this is clearly a good safeguard for consumers but it has also been said that it can work against development of innovative technologies.  To counteract that, Italy, France and the UK have the option of allowing energy suppliers to deploy innovative technologies and to monitor the resulting energy savings which can subsequently be claimed.  This option has been rarely used and the UK is currently looking at how innovation can be encouraged better by other means from 2008-2011.  However, many studies have shown that at least 20% of energy consumption can be saved through the widespread application of existing, proven technologies, which is clearly the short term policy focus.

To date, most of the obligations have focussed on energy saving measures.  This means that co-generation, solar water heating and other renewable forms of heating have generally been included but there has been less promotion of renewable energy generation technologies.  This is not least because there are usually different policy mechanisms to support the development of renewable generation technologies.  However, in the UK, with the move to CO2 targets in 2008, supporting household and community scale renewable energy sources will be possible under the new obligation.

Although the countries have various end-use sectors to which the obligations apply, they all have included the residential sector. While lighting and heating measures feature strongly in all four countries, there are marked differences in the appliance and insulation areas (Table 3.6).  Some of this can be explained by local conditions, e.g. the UK has low standards of insulation in its existing stock.  

At present there are few, if any, energy saving credits for "behavioural" change measures such as smart meters with consumption feedback to households, energy efficiency advice, etc.  However, this reflects more the difficulty in establishing firm energy saving values and the appropriate lifetime for such measures rather than any fundamental barriers.  Indeed, work is currently underway in the UK to look at introducing energy savings for such measures from 2008.

Monitoring and Verifying Energy Savings

The great majority of projects have been carried out by utilities utilising the estimates or extant energy savings, or in the case of industrial and commercial measures, by scaling engineering estimates of proven energy savings .  For example, in the residential sector although the energy savings will vary from household to household, for either insulation or new appliance. Measures are being adopted, as long as the average energy saving has been established. Then the use of an estimate or extant savings will represent the real situation because of the large numbers of households involved. This approach greatly simplifies the monitoring and verification process which in effect becomes the same as counting the number of energy efficiency measures implemented and can be verified using the standard "dip check" or random sampling audits. In addition, this method has low costs for implementing and verifying such energy efficiency measures. 

In the UK, energy suppliers submit outlines in advance of what they intend to do and the energy savings they are likely to claim.  This has benefits both for the Regulator and the energy company in minimising later disputes in terms of energy savings achieved.  The final approved energy savings are of course related to the actual outcome rather than the outline.  In the fully traded White Certificate schemes, as in Italy, the energy company can either provide the approved savings from its own projects or purchase an appropriate number of White Certificates to meet their targets .

Obviously, to make the estimated energy saving or engineering estimate approach work successfully, transparent and clear information should be available from the body responsible for accrediting energy saving values or White Certificates to energy companies, which are published well in advance of the duration of the obligation.

The Treatment of "Free Riders"/Additionality

As with all public policies, there is a need to address dead-weight/free riders/additionality issues i.e. end users that would have utilised the energy saving measure anyway.  The way this issue is handled is linked to the size of the target but this issue is topical for French, Italian and UK energy efficiency obligations.

For low targets, when the size of the activity being supported by energy companies is comparable with the historic activity in the energy saving measure, minimising the number of free riders on a project-by-project basis has been widely used . 

However, as the scale of the activity being supported by energy companies becomes greater than the earlier activity in the energy saving measure, a different approach is required particularly for measures using retailers to reach the energy companies' customers.  This is currently the situation in the UK and the approach taken there has been to build the free riders into the target .  In other words, the Government decides on the energy savings over "business as usual" target it wants to achieve and then adds on the business as usual activity to determine the final target for the energy company.

Supplier or Distribution Obligation.  Which is Optimum?

There are pros and cons of both approaches.  In favour of suppliers is their strong links to the customers and, in a liberalised market, increasing marketing skills.  In favour of a distribution levy is the fact that utilities are more stable organisations, being often regional monopolies, and they are already regulated.  The main disadvantages are that distribution companies' contacts with customers, particularly in the residential sector, are usually only when there is a failure in the wires or pipes business and that depending on the Government's view, such distribution levies can be counted as part of public expenditure, a cost which Governments usually wish to minimise.

The key is that both systems can be made to work but certainly there is more of an incentive for energy suppliers to reduce the amount of energy supplied to their customers than for distribution companies.

Meeting the Target: Energy and CO2 savings

The experience to date in the UK, Italy and Flanders has shown that the companies have easily met their targets.  In the UK, on average the energy suppliers have met their targets with an expenditure of 20% less than the Government expected, and since 2002 have carried forward energy savings from one phase of the obligation to the next.  In Flanders, the energy distributors met their targets at 40% less cost than budgeted in 2004 and at 24% less cost than budgeted in 2005.

In Italy, the first year's targets have been met and the corresponding price of White Certificates indicates that the distribution companies have met their targets well below the allowance in the distribution price.

In Flanders the energy and carbon savings are for electricity users only.  By the end of 2005, the electricity distributors had reached 494GWh/year electricity savings with their larger customers (>1kV) and 520/yearGWh with low voltage customers (mainly residential); both figures are inclusive of "deadweight".  In principle, this would correspond to over 6% of Flemish electricity consumption .  The reported CO2 savings are 0.37Mt/year (end of 2004).

In the UK, the energy savings from the first period (2002-2005) have been estimated to be around 40TWh for electricity and 100TWh for fossil fuels with both figures being lifetime savings including deadweight .  The CO2 savings have been estimated by Defra after correction for deadweight to be 0.5Mt/year (1.8MtCO2/year) or 1.3% of household annual emissions.

In Italy the targets for 2005 were equivalent to 456GWh/year electricity saving and 675GWh/year gas savings (both figures in delivered units).  The Italian regulator has reported that both figures were easily exceeded.  The 2005 annual target corresponded to 0.15% and 0.13% of Italian electricity and gas consumption respectively.  This is only the first year's savings and the targets will increase in subsequent years. In the first year, the CO2 savings were 0.76 Mt/year.  To date, the CO2 savings are 2.3 Mt/year.

Targets were generally set at a low level to initiate the process, which explains why they were reached in all countries. They consisted so far of simple measures and have not yet tapped the large energy efficiency potentials in the structure of the existing buildings .

Trading

To date, the experience of trading in White Certificates has been somewhat limited. This is perhaps to be expected because only in the more recent Energy Efficiency Obligations in Italy (since March 2006) and France there are opportunities for market players other than the energy companies to attain independently and trade White Certificates.  Thus the market is in its early stages and will undoubtedly grow over time.

In the UK, trading is only permitted between energy suppliers and has rarely been used.  The main example of trading was between some of the smaller companies which had their requirements fulfilled by one of the six major energy suppliers.

In Italy, because of the very limited trading nature (less than 20% of the White Certificates in 3 months were traded), it is not entirely clear what the prices of the White Certificates in the market place are signalling.  Most of the White Certificates in Italy have been obtained directly by the energy distributors working either in bilateral or contractual arrangements with deliverers of energy efficiency measures .

The current (May 2007) situation in Italy is that the average price of White Certificates is around €43/toe primary energy for electricity and around  €85/toe for gas.  Both are well below the €100/toe primary energy allowed in the distribution price control.

Although, trading is not widely permitted in the Flemish and UK, certain facets are similar.  For example, energy companies may carry forward excess energy savings from one phase to the next.

Financial Costs and Benefits Arising from Energy Efficiency Obligations

Although all the costs are ultimately borne by energy users, it is perhaps helpful to get a range of the transaction costs involved and who initially pays them.  Currently, only the UK has the full data publicly available (Table 3.7).

The great majority of the total cost is associated with promoting and installing the energy saving measures.  Although energy suppliers are the single biggest contributors to these costs, money is also coming from those customers that can afford to pay as well as landlords, charities, manufacturers etc.  The energy companies need to spend significant sums to ensure that these energy saving measures reach their customers: such energy supplier indirect costs were evaluated in 2002 to be around 20% of total costs .

Then there is the cost of monitoring, verification and general administration of the energy saving projects by the energy suppliers. In the UK, the energy regulator Ofgem carries out this task and estimates that this cost is less than 1% of energy supplier expenditure.  Finally, there is a cost of setting the target and the associated legislation .

Data on costs and benefits are available for Flanders, Italy and the UK.  They are evaluated from an energy company's perspective and also from a national perspective.  The former only considers expenditure by the energy supplier; the latter includes the costs to all participants, i.e. energy companies, customers, third parties (e.g. local authorities, landlords, manufacturers, charities etc.).

The energy company perspective

Comparison of cost effectiveness from an energy company perspective is complicated by the difference between the targets (annual or lifetime), the use of a discount rate (6%) for lifetime savings and the different measure mix and hence average duration (Table 3.8).  The Flemish value is the average of residential and non-residential customers.  Separating the two classes, the values are for non- residential 2.7c€/kWh and for residential 7.6c€/kWh (with both figures being annual savings compared to a 0.027€/kWh "one-off" cost of the measures).

To enable a rough comparison of all three countries, we have used an average life of 12 years and undiscounted savings.  The annual cost per unit of electricity in the UK is estimated at around 6.8 c€/kWh.  As the UK obligation is only targeted at households (more expensive than saving energy in commercial and industrial sectors) then the results are broadly comparable. 

The estimate for Italy is derived using a market price for White Certificates for electricity of €70/toe primary energy saving; this is likely to be an overestimate as the cost of a White Certificate is currently below €50.   In France, the expectation is that the cost to the energy suppliers of saving a unit of delivered energy will be less than 1c€/kWh lifetime.

The national perspective

Insufficient data have been published for Flanders to make any estimate possible; likewise for Italy.  The only readily available data are from the UK: results from the evaluation of the energy efficiency obligations are drawn upon here .

The traditional way is to look at the lifetime net present value (NPV) of the obligations, which ran for 3 years from 2002-2005.  The responsible UK Government Department, Defra, has evaluated that for a total investment (i.e. all costs of the energy suppliers, customers and third parties) of €0.9billion, the NPV (using a 3.5% discount rate) is €4.5billion.  In terms of resource cost to save a tonne of CO2, Defra evaluated this at - €450/tC, i.e. a net benefit rather than a cost.

An alternative way of looking at this in more familiar units is to look at the cost to the nation of electricity or gas saving: 2c€/kWh for electricity and 0.7 c€/kWh for natural gas; both figures are considerably less than the average consumer prices during that period of 10c€/kWh and 2.6c€/kWh respectively.