1.2 Functions of indicators and indices

The usefulness of indicators can be increased by putting them in an appropriate context or framework. The most widely accepted of the many proposals is the so-called Pressure-State-Response (PSR) model, which divides indicators in three categories as follows:

The PSR model was developed in the 1970s by the Canadian statistician Anthony Friend, and subsequently adopted by the OECD’s State of the Environment (SOE) group. The European Commission’s indicator development follows this framework. Some organisations prefer variants of the PSR model; for example, the UN Commission for Sustainable Development (UNCSD) bases its indicator set on the Driving force-State-Response model (DSR) model, which allows for a better inclusion of non-environmental variables.

1.2.2 The Driving force-Pressure-State-Impact-Response (DPSIR) model

For practical purposes, and in particular for the goals described in the Green Accounting Communication, the PSR model is sufficient. However, for compatibility reasons (e.g. to the DSR model), and for a better description of underlying economic trends, the indicator community has formulated the Driving force-Pressure-State-Impact-Response model, which includes P-S-R as special cases. Please note that, as explained before, the UN CSD indicator process uses the DSR model, where the term “Driving Forces” is used synonimous for “Pressure” [1].

P Pressure indicators describe the variables which directly cause (or may cause) environmental problems. Examples: toxic emissions, CO ₂ emissions, noise etc. caused by road traffic; the parking space required by cars; the amount of waste produced by scrap cars.

I Impact indicators describe the ultimate effects of changes of state. Example: the percentage of children suffering from lead-induced health problems; the mortality due to noise-induced heart attacks; the number of people starving due to climate-change induced crop losses.

R Response indicators demonstrate the efforts of society (i.e. politicians, decision-makers) to solve the problems. Examples: the percentage of cars with catalytic converters; maximum allowed noise levels for cars; the price level of gasoline; the revenue coming from pollution levies; the budget spent for solar energy research.

Eurostat focuses on Driving force (e.g. sectoral trends), Pressure and Response indicators, and on linking such indicators to standard socio-economic statistics. Complementary to this effort, the European Environment Agency (EEA) will concentrate on state and impact indicators, and on a comprehensive description of the full PSR chain.

1.2.3 Main properties and functions of indicators in the DPSIR framework

D Driving force indicators are not very responsive ("elastic"): the monitored phenomena, e.g. road traffic, are driven by powerful economic forces, and therefore it can hardly be expected that these trends will change drastically in future. For example, politicians cannot seriously suggest to abolish private cars, if they want to stay in office. However, Driving force indicators are useful to:

P Pressure indicators point directly at the causes of problems. One specific feature of pressure indicators is that they should be responsive, that is, a decision-maker has indeed a chance to reduce the indicator (and thus the problem) by launching appropriate actions. They will also serve as an incentive for rational solutions, since they demonstrate the effectiveness of political action early enough to hold responsible those who launched the action.

S State indicators, in contrast, are often too slow. For example, a state indicator showing the acidity of forest soils points back to the NO _x and SO ₂ emissions of the last ten years; the politically responsible persons may have retired in the meantime. On the other hand, state indicators can serve to make a first assessment of the situation (what is the current state of the forest soils? where could corrective measures be applied?) , and they are certainly appropriate tools to plan habitat restoration and similar clean-up activities.

I Impact indicators react even slower than state indicators. When the impacts are felt, it is usually too late for action. In addition, it is rarely possible to establish solid statistical correlations between pressures, state, and impacts, due to the enormous delays and the influence of non-environmental variables. The main purpose of impact indicators is to demonstrate DPSIR patterns, in particular: cause-effect chains , and to facilitate informed discussions about actions to avoid negative impacts in future. In this sense, they are not statistical "indicators", but scientific "decision models".

R Response indicators are very fast, since they monitor the measures which are intended to make the slow socio-economic system move. Examples: 1. rising energy prices due to the introduction of an energy tax can be observed immediatey, while the full effects of this measure (decreasing energy use and CO ₂ emissions due to behavioural, technological and other adjustments [2]) will be noted only five to ten years later; 2. the volume of money spent by public authorities and industry for environmental protection measures can serve as a quick indication whether appropriate actions have been launched. There is no a priori guarantee, however, that political responses (actions, measures, instruments, budget increases, ...) will be useful and efficient; the monitoring of success can be performed only through pressure and state indicators.

1.2.4 Data, figures, indicators: some basic definitions

- Data vs. statistics: Data are figures that need further processing (e.g. aggregation to national level, adjustment for season, climate, economic cycles etc.), before they can be called statistics.

- Academic vs. statistical figures: In principle, statistics coming from official sources, e.g. the national Statistical Services, Eurostat, OECD or the European Environment Agency, enjoy more confidence than academic figures obtained from specific studies. Users assume that official figures have been produced with standardized definitions and reflect “mainstream thinking”, and they know that, on the other hand, scientists may produce a wide range of results depending on differences in basic assumptions and definitions.

- Statistics vs. indicators: Statistics are figures describing real phenomena according to an exact definition. In practice, statistics often need interpretation, and sometimes they have footnotes attached which explain why the figures as such are correct but may not look plausible at first sight. Indicators, in contrast, should send correct messages without a need for further interpretation. Indicators may require adjustments, e.g. for seasonality or climate, and often they are related to interesting reference variables, e.g. CO ₂ emissions per capita , fertilizer use per ha of arable land .

- Reports vs. indicators: Especially for a complex policy area like environment, detailed reports (e.g. the EEA’s Dobris Assessment, World Resources) are an indispensable tool for experts. Statistical figures are often embedded in the text and serve as the basis for an in-depth discussion of policy priorities and options. In contrast, pure indicators are “executive summaries” addressed to non-experts who want to get a quick impression of basic trends without the need for further interpretation. For example, a proper description of trends in the use of water resources may occupy twenty pages of text, maps and graphs in a State-of-the-Environment report; the journalist who has to write about the same subject will be granted a 10 by 10 cm space, and thus is forced to rely on highly condensed information, e.g. a handful of key indicators.

- Indicators vs. indices: Sometimes, the term “index” is used for indicators related to a baseline year (“Index 1990=100”). In the context of environmental indicators, however, “ index” means an aggregation of indicators with similar impacts . For example, greenhouse gas emissions such as CO ₂, methane, nitrous oxide, CFCs could be condensed into one greenhouse gas index showing the total contribution to Climate Change caused by these gases (based on the common unit Greenhouse Warming Potentials ). Main purpose of such aggregations is to communicate detailed information to an audience that requires condensed, “simplified” information.

1.2.5 From raw data to indicators: the Information Iceberg

The iceberg picture stresses that most of the efforts needed for producing good indicators and indices happen “under the surface”. There have been many workshops on “indicators of Sustainable Development” in the last five years, but almost no “public” events to discuss the quality and coverage of the statistics needed to implement the various brillant concepts produced by indicator experts throughout the world. This is particularly regrettable since data collection needs time; between the formulation of a demand for data and the first delivery of figures based on a reliable statistical data collection, five to ten years may have passed. Ad hoc estimates may serve to provide a first picture how the final indicator could look like, and to test the validity of the concept, but such estimates are usually not robust enough for the controversial debates of environmental politics.

1.2.6 Indicators: neutral “tools”, or “ammunition”?

There is an overwhelming amount of information on the environment available. Thousands of scientific reports present figures that describe relevant phenomena in quantitative terms. What these figures have in common is that they are different. The old joke “ask three scientists, and you'll get four different opinions” is not really a joke. The lack of standardization and harmonization would prevent us anyway from calling such figures “statistics” or “indicators”. However, in addition to such purely technical considerations, it is evident that too often the differences between scientific figures become more plausible when looking at who has sponsored the research. It is very unlikely that two groups of researchers, one sponsored by an environmental lobby, the other by an industry or agriculture association, come to similar results. Obviously, many “scientific” figures are being produced as arguments against the opposite party in environmental policy.

For the Commission's indicator working group, the answer was clear from the very beginning. Environmental Pressure Indices, as suggested in the “Green Accounting Communication”, must be neutral if they are to be accepted by the main players in environmental policy. One practical step to achieve this was to make the indicator development transparent by consulting many scientists (see Chapter 1.5, The Scientific Advisory Groups (SAG) surveys). On a more political level, it was a logical decision to put the indicator development into the hands of the statistical services of EU Member States (coordinated by Eurostat), since political neutrality must be an obvious working principle for statistical services.

1.2.6.1 Rules derived from the “neutral tools” principle

The main components of “welfare politics”, economic, environmental and social policy, can be disaggregated to sub-clusters.

Environmental policy is structured along policy fields like “ Air Pollution ”, “ Climate Change ” or “ Waste”, which in turn can be further disaggregated to sub-fields such as “hazardous” or “municipal” waste.

Indicators should closely follow this structure, so that at each level of the hierarchy the political process disposes of comparable figures.

Of course, many scientists will claim that there are “better” structures for environmental policy. However, they must be introduced by the political forces, not by indicator developers or statisticians.

Not by accident, this figure resembles in many aspects the “information iceberg” - the more detailed the policy questions, the more detail is required for the figures.

1.2.6.2 Transparent debates and informed decisions

The complex character of environmental policy debates often gives the citizen no chance to be involved in the decision-making process. Who is able to judge whether it is better to dump household waste on landfill sites, or to burn it in waste incineration plants? Such options are relevant for ordinary citizens who will have to live near the waste disposal facility, they are being controversially discussed (e.g. in the local newspapers), but the citizen cannot really participate. Flooded with technical details, they can only guess the truth by listening to the experts, and trying to guess which side is more credible in their arguments.

A system of indicators structured along the lines shown in Figure 4 above might at least enable the citizen to compare the experts’ arguments more systematically. From the viewpoint of democracy, that would be an improvement. But would it also be in the interest of the societal actors? Economic sectors, environmental NGOs and politicians may feel uneasy to discuss their arguments with a wider audience, but in the long run all sides can profit from commitments that are based on a public consensus, and thus cannot be easily overthrown by sudden switches of public opinion. Informed decisions are more binding than those made behind closed doors, and investments made on the basis of hard but public compromises are well-protected against the “chemical of the month” phenomenon - and therefore safer.

1.2.6.3 Abstraction and Transparency

Using the hierarchical structure shown in Figure 4 for an indicator system implies aggregation (see Chapter 1.4, Aggregation and linkages ). There is a trade-off between the need to simplify and condense information (making political mechanisms more transparent), and the desire to keep the indicators so simple that their users can still link them to reality. For example, an “air pollution index” will be more abstract than a “particle emissions indicator”, and might therefore be more difficult to understand for the users. On the other hand, the public would not profit from e.g. a set of six air pollution indicators, especially if some of them point upwards, others downwards.

The experience with other complex issues, such as the economy, shows that condensation and abstraction is necessary. Nobody wants to see the thousands of time series that are necessary to create GDP. In spite of the high degree of abstraction, GDP is accepted as an indicator because a) there are no alternatives for measuring economic performance and b) the methodology is accepted by all relevant actors, from industry associations to trade unions.

Likewise, environmental indices could be acceptable to users if all relevant actors could reach a consensus on the calculation methodology. At present, only the Climate Change expert community has reached a consensus on the Greenhouse Warming Potentials (GWP). A “greenhouse gas emissions index” is more abstract than “CO ₂ emissions”, but offers the potential to discuss all contributions to Climate Change in a comparable way, and to make sure that the other greenhouse gases (e.g. methane emissions of agriculture) are not “forgotten” in the political process.