Modelling the Relationship Between Energy-Saving Behaviour of Industrial Companies And

Modelling the relationship between energy-saving behaviour of industrial companies and the role of environmental policy instruments

Prepared for presentation at the Open Meeting of the Global Environmental Change Research Community, Rio de Janeiro, October 6-8, 2001

Keywords: energy-saving, industrial companies, determinants of investment behaviour, characteristics of policy instruments, effectiveness of policy instruments.

J.J. van Wijk, R.F.J.M. Engelen, J.P.M. Ros

National Institute for Public Health and the Environment (RIVM)

Laboratory for Waste Materials and Emissions (LAE)

PO Box 1, 3720 BA Bilthoven, The Netherlands

Webpage: www.rivm.nl

E-mail:

tel: +31 30 274 35 73

fax: +31 30 274 44 17

1. Introduction

Industrial energy savings and the influence of policy instruments have received considerable attention within the context of the Kyoto Protocol. The Dutch government employs several policy instruments, like voluntary agreements and permits, to stimulate industrial companies to adopt energy-saving techniques. The crucial question is whether these instruments, and more specifically, whether the design of the instruments is such that sufficient pressure can be wielded on companies to implement energy-saving techniques and thus bring the national climate goals within reach. This question concerns both diagnosis and prognosis. In the diagnosis, explanations for emission trends, as derived from monitoring results, are sought, while in the prognosis, analyses are carried out to estimate the feasibility of quantitative reduction goals.

For the necessary insight into industrial energy savings and the (effective) influence of policy instruments - for both the past and the future - the Netherlands Institute for Public Health and the Environment (RIVM) has, in co-operation with other institutes, developed MEI-Energy, the Model Effectiveness of Policy Instruments for Energy-Saving in Industry (see Elzenga et al. 2000; Van Wijk et al. 2001). This computer model simulates the investment behaviour related to energy-saving techniques of Dutch companies within a particular industrial sector. The model is meant to be an ‘expert-supporting system’, guiding experts in translating their knowledge of a certain case into the diagnosis and prognosis of the implementation path of a technology. For the moment, the model is operational but still in the validation stage.

Here, we will focus on the scientific policy-related basis of the model. Readers interested in the technical details are referred to background documents (De Groot et al. 2001, Engelen et al. 2001). Section 2 introduces the model structure, while the model component simulating the non-economic decision-making process is elaborated in section 3. Section 4 describes the role of policy instruments in this decision-making process. The first output data illustrating how policy scenarios are analysed with the model are presented in section 5. Finally, conclusions are drawn in section 6.

2. The model structure and purpose

Empirical studies, for example, by Gillisen et al. (1995), De Groot et al. (1999) and Velthuijsen (1995), show that industrial companies are stimulated to invest in energy-saving techniques under the influence of several factors, originating in the following four clusters:

(1)  technical energy-saving measures (e.g. complexity, investment costs, technical depreciation time);

(2)  the industrial sector (e.g. the age of a current installation complex within a sector, level of organisation, innovation efforts and attitude towards the environment);

(3)  social surroundings (e.g. media focus on the greenhouse effect, actions of environmental groups);

(4)  policy instruments and policy environment (e.g. taxes, permits, voluntary agreements, enforcement activities).

Strongly guided by these factors, companies decide whether they will or will not (yet) implement an energy-saving technique. A diffusion curve for each technique belonging to a certain sector can be constructed yearly using the ICARUS[i] database (Beer et al. 1994; Alsema et al. 2000). In MEI-Energy, two kinds of simultaneously occurring decision-making processes are combined (see below):

(1)  an economic-rational decision-making process based on quantitative criteria such as energy prices, depreciation period of the technique, investment costs and financial policy instruments like taxes and grants etc.;

(2)  a driving-forces qualitative decision-making process in which factors like the sector’s attitude towards energy-saving, the sector’s expectations with respect to the influence of the technique on the product quality and public pressure on the sector all play an important role. These factors are represented by seven driving forces, which will be elaborated below.

As previously mentioned, the MEI-Energy model calculates the diffusion curve per technique per sector on a yearly basis. Figure 1 presents such a curve and shows it to consist essentially of three parameters: the preparation time for implementing the technique (tv), the speed of diffusion (Sp) and the part of the sector that will finally implement the technique (at Pt). The processes in (1) and (2) are incorporated into modules. The economic-rational decision module determines the largest component of minimal energy-saving, after which the driving forces module may generate more energy-saving.

Since each technique has a characteristic energy-saving factor, the total energy savings of a sector can be computed in relation to the business-as-usual scenario as the point of reference. The effectiveness of policy instruments can be calculated by comparing energy-saving effects of sets of techniques.

The addition of the driving forces module to a more standard economic perspective on technology diffusion, as often used in models (e.g. the Dutch models NEMO and SAVE), is seen as an appealing step in coming to a formal representation of the factors influencing the investment behaviour of companies in energy-efficient technology (De Groot et al. 2001:18).

Figure 1 Diffusion curve.

3. The driving forces module

From the previously mentioned four clusters of determinants for investment in energy conservation, seven driving forces have been defined and outlined in Table 1[ii].

Table 1. Description of driving forces within the MEI-Energy model

Driving force / Description
1.  Financial-economic situation / The extent to which the sector has financial possibilities for investment and to which there will be negative socio-economic consequences.
2.  Level of knowledge / The amount of knowledge the sector possesses on energy-efficient technological options.
3.  Complexity of the technique / The degree to which the energy-saving technique imposes technical barriers to implement it and keep it operational.
4.  Market pressure / The degree to which the technique positively or negatively influences the production process and the product quality.
5.  Public pressure / The degree to which non-market parties (like environmental movements, shareholders and the public) put pressure on companies within a sector to improve their energy performance.
6.  Attitude towards the environment and energy / The degree of willingness to be expected within a sector to invest in energy-saving.

7.  Policy pressure

/ The pressure that the government puts on companies to improve their energy performance by setting (behavioural) standards through policy instruments and the enforcement of these standards.

The driving forces should be seen as stimuli that industrial companies receive[iii], forming incentives or barriers for them to implement an energy-saving technique. This fits traditional decision-making models: the relevant factors are argued (and calculated) from the point of view of the actor making the decision[iv]. Although MEI-Energy simulates the behaviour of one group of actors (namely, all companies of an industrial sector), the driving forces represent the behaviour of other actors (e.g. environmental and consumer groups generate public pressure, while an industrial society plays an important role in the attitude of a sector[v]). Although one driving force is defined by environmental policy instruments (policy pressure), policy instruments occupy a central position in other driving forces as well.

Driving forces are seen to be independent. In practice, this is, of course, not the case. If an issue does not receive attention in political circles, environmental groups often try to put the issue on public and political agendas. In other words, there is a link between policy and public pressure. There are also more multivariate relationships possible between driving forces. To keep the model transparent, we have not modelled interactions of driving forces. The user of the model should, however, take the interaction of driving forces into consideration when scoring the influence of driving forces[vi].

4. Policy instruments in the model

4.1 Characteristics of policy instruments

In general, policy instruments are classified into three types: 1) command and control instruments (legislation, permits); (2) economic instruments (taxes and subsidies) and (3) communicative instruments (public relations, education etc.). The position of voluntary agreements within this classification is ambiguous. Voluntary agreements can be seen as a stimulative variant of the command and control instruments (Driessen and Glasbergen 2000:54). Agreements are often accompanied by subsidies; this means that the agreements can also be placed alongside the economic instruments. In carrying out the Dutch energy-saving agreements, the intermediary organisation, the Netherlands Agency for Energy and Environment (NOVEM), has played a major role in supporting companies by supplying knowledge on energy-saving. Taking this perspective will allow an agreement to be used along with communicative instruments.

Instead of classifying policy instruments into one of the three above-mentioned types, they can be characterised in another way. For example, a voluntary agreement contributes, in general, positively to the attitude of a sector since the sector has been involved in the design of the agreement. Besides, companies that fall under an agreement tend to organise themselves in working groups, which helps to improve the level of knowledge in the sector (Rietbergen et al. 2000:81). Multi-laterality and knowledge support thus represent one of the characteristics of a voluntary agreement valued by a sector. In conclusion, the use of the traditional distinction between the stick, the carrot and the sermon, i.e. the respective legal, economic and communicative instruments (compare to Winsemius 1986), seems inadequate for classifying policy instruments. Instrument characteristics would seem to offer more possibilities for providing insight into the significance of policy instruments in the (investment) behaviour of companies.

Instrument characteristics can be derived from Dutch political and social environmental sciences. Since the mid-eighties numerous studies on the effectiveness of environmental policy, largely from an instrumentalist perspective, have been carried out. In line with this ex-post-evaluation research, studies have tended towards ex-ante assessments of policy instruments. The key issue here is determining the different contexts and organisational settings under which policy instruments are effective (Leroy and Nelissen 1999:79-83). Characteristics of policy instruments have been identified within this instrumentation research. A characteristic of an instrument can be seen as a specific part of the policy instrument design, so as to stimulate a company to act according to the behavioural standard imposed by this instrument upon it. The table below overviews the instrument characteristics incorporated into the MEI-Energy model.

Table 2. Overview of MEI-Energy instrument characteristics

Instrument characteristic / Description
1.  Ambition level / The degree to which a change of existing behaviour is demanded from companies (distance to target).
2.  Multilateral versus unilateral / The degree to which the sector has been involved in the design of the policy instrument. A multilateral instrument is the result of consultation between government and target group (and possible other stakeholders). A unilateral instrument is defined without such a dialogue (compare Ligteringen 1999:100; Wit et al. 1999:6).
3. Fairness (national/international) / The degree to which the instrument reflects the principle ‘what's sauce for the goose is sauce for the gander’ within an international and national context.
4.  Supply / withdrawal of resources (financial and knowledge) / The degree to which companies within a sector receive know--how and/or financial support to observe the behavioural standards as stipulated in the policy instrument.
5.  Supply / withdrawal of resources (awareness of the environmental problem) / The degree to which activities of research centres and environmental movements on the climate problem are financially supported by the government. These activities contribute to national environmental awareness, stimulating companies to develop a sense of urgency for acting environment-friendly (Dieleman 1999:82-83).
6.  Monitorability / The degree to which non-compliance to environmental standards is monitorable by the policy instrument implementer.
7.  Scope / The focus of the instrument: only energy-saving or focused on more issues e.g. de-materialisation.
8.  Legal force / The degree to which companies are punished when they either do not, or do not sufficiently, employ the instrument (jurisprudence).
9.  Level of execution and enforcement / The degree to which companies are monitored by public bodies in the employment of the instruments. Important is the level of knowledge within public bodies and their human resources capacity.
10.  Publicity / The issue of confidentiality: do public bodies treat individual monitored figures from companies confidentially or not ?
11.  Imperativeness / The degree to which the instrument limits the freedom of companies within the sector.
12.  Range / The degree to which the instrument applies to the companies in a sector.

4.2 Policy instrument characteristics and driving forces

In the preparation of the Fourth National Environmental Policy Plan (NMP4), the Dutch Minister of the Environment questioned the use of voluntary agreements, which, ever since the nineties, have formed a key instrument in Dutch environmental policy. The reason for the Minister’s doubting the effectiveness of agreements is that these instruments have not been successful in all policy domains. The Dutch Federation of Industry counter this, claiming that more than a hundred of the agreements work very well (NRC Handelsblad, 21/5/01).

The NMP4 itself has stated environmental agreements as being partly successful. Crucial to the success of an agreement is the commitment of involved parties to take their responsibility. For example, industrial associations should rap free-riders over the knuckles, while the government should enforce the agreement. The report concludes that the use of agreements should be closely scrutinised: is an agreement the right instrument to realise an environmental goal?[vii] If so, agreements should be designed, carried out and enforced in such manner that realisation of the intended goals is guaranteed (VROM 2001:73).

An interesting issue in this debate on agreements concerns the question: Would environmental goals have been reached if the unsuccessful agreements had been stricter in their design (i.e. showing a higher ambition level, easier to control, fewer behavioural options etc.) and thus capable of imposing more pressure on companies? To answer such a question on effectiveness using MEI-Energy, we need a connection between policy instrument characteristics and driving forces (see Table 3). This connection will clarify the significance of policy instruments in the decision-making process of companies.