27.7.2007   

EN

Official Journal of the European Union

C 175/21

2.2.1

Logically enough, DG ‘Education and Culture’ and DG ‘Employment, Social Affairs and Equal Opportunities’ include social aspects in their IAs. Furthermore, the ‘principle of proportionate analysis’  (7) induces differences in the degree of integration of social elements in other areas. This raises the question of whether or not social aspects (including items related to the EU Social Agenda) need be considered as a bottom line and also how to deal with proposals not related to social issues or likely to have a weak social consequence. Empirically, the study of the Istituto per la ricerca sociale (see e.g. footnote 2) reveals that ‘IAs lacking consideration for social aspects tend to concentrate in the economic field. Here, one third of the IAs includes social aspects only marginally or not at all’ (8).

2.2.2

Obviously, ‘when the social relevance of the measure is self-evident (…), social aspects are widely considered and quite developed throughout the impact assessment document’ (9). There, ‘the employment issue clearly emerges as the most recurrent and stressed social consequence’ (10).

2.2.3

According to the Istituto per la ricerca sociale, ‘the degree to which social elements are considered is not necessarily “proportionate (…)” to the policy content and its likely impact. (…) In many cases these impacts are only described generically (…)’ and ‘are based on (…) shared assumptions (…). Such relations are seldom discussed when taking into account the specific content of the measure, the target population and interested territorial areas, the specific choice of policy instruments and the effect of the implementation process’ (11). The study carried out by this institute also states that ‘several IAs have not envisaged any correlation with other policy domains or EU policies’. While the work involved in conducting impact analyses is considerable, the latter must not be incomplete or superficial, or their shortcomings will undermine the value of the legislation.

2.3.1

Evaluation of the impact of a legislative draft is not a ‘tick box activity’. It also needs to be monitored — ideally by or in close cooperation with users of the law, especially those who will be most directly concerned. Since the social dimension is one of the three assessment criteria for EU policies there is a need to organise a standard procedure — both transparent and straightforward — for the collection of targeted inputs in the context of IAs. A few options can be mentioned:

3.1

A number of questions must be posed in order to determine what methodology could be recommended:

4.1

In view of the complexity and importance of evaluating the social consequences of legislative proposals, all actors concerned, i.e. social partners as well as representatives of civil society organisations, should reflect upon the following methodological issues:

4.2

A public hearing at the Committee has given social NGOs, social partners, other actors of organised civil society and experts a chance to express their views and discuss the draft opinion so as to pass clear messages to the European institutions in general and to the Commission in particular.

4.3

Ultimately and in view of the fact that social IAs are an element of paramount importance in the EU decision-making process, the Committee should make proposals on how to bring about improvements and better integrate civil society organisations in this procedure.

5.1

Several different social indicator systems already exist at national and international level, but it is necessary to ascertain how useful and how suitable they would be for the specific purposes of impact analysis.

5.2

Social indicators have been developed in a number of countries over the past thirty years as an alternative to assessing the appropriateness and impact of economic policies in terms of basic, quantitative data, with a view to ‘steering’ social development in parallel with economic development and developing the capacity to measure social well-being and its evolution.

5.3

As a result, there has been a significant growth in the amount of social statistics available, particularly in major areas of collective responsibility such as education, health, social protection, the environment, housing, transport, research and unemployment. However, unless this data is organised, summarised and interpreted, it does not automatically lead to the creation of social indicators.

5.4

An indicator is ‘merely a statistic, to which particular importance is attached for the purposes of knowledge, judgement and/or action’ (12). As far as impact analyses are concerned, it is not simply a question of compiling social statistics, from various sources, for each country, but of structuring this data in order to assess the current situation in relation to specific themes, selected on the basis of their relevance to the impact analysis.

5.5

It is quite possible that in some areas, only sparse or fragmented studies and surveys may be available, making it impossible to conduct a cost/benefit analysis. For example, we already know that certain categories of pesticides are harmful to health and that concentrations in excess of a certain threshold cause serious illness. Yet, although a decision to reduce the use of chemical pesticides would have a positive impact on the health of the general population and of workers exposed to these pesticides, it would be impossible to produce exact figures on the long term benefits of this policy within the framework of a proportional impact study.

5.6

On the other hand, the social and health aspects would clearly justify the proposed measure and would provide considerable reinforcement for the economic arguments (which would include the reduction of production costs for agriculture and the resulting increase in competitiveness). In addition, the fundamental right to a clean environment could be raised in support of the proposal.

5.7

In practice, we now have access to a considerable volume of social statistics and these have also diversified in accordance with the changing foci of public debate in the various countries (such as working conditions, employment of young people, older people and women, crime, income inequalities, discrimination at work and relocations). Further, although few social indicators were drawn from this mass of statistics until very recently, fortunately, in the new socio-economic climate which has developed over the past ten years, where emphasis is again being placed on the role of the state in social policy and regulating the market, their relevance is once again being acknowledged.

5.8

However, social indicators do not always serve any great purpose if they stand alone, and take on more meaning if they are integrated into a broader concept, such as social and economic development or sustainable development. They now come from an increasingly diverse range of sources, not just central government but also from NGOs, and the think tanks operated by major foundations. They are also being presented in increasingly diverse ways, ranging from the selection of statistics to thematic surveys, and the compilation of data to form composite thematic or general indicators.

5.9

A number of international bodies publish social indicators and statistics and conduct comparisons between their member states. In this connection, the main bodies of relevance for the EU countries are (in no particular order) the OECD, the UNDP, the European Commission, particularly Eurostat, UNESCO, the World Bank and the ILO.

5.10

This very diversity of sources poses a number of problems in itself: firstly, with regard to the quality of the statistics, in that not all countries have sophisticated Statistics Offices, secondly in terms of comparability and thirdly in terms of harmonising concepts. For, ‘the choice of indicators for measuring the social convergence of the EU countries is a highly political issue. The indicators used for comparisons are not neutral: they not only reflect priorities but also sometimes particular understandings about what kind of society is desirable, which may legitimately vary from one state to another. The example of unemployment demonstrates that some indicators may have tangible and possibly even negative impacts on policy orientation (13). Notwithstanding, as things currently stand, the construction of indicator systems is left entirely to technicians.’ (14)

5.11

The use of GDP and growth as indicators of social wellbeing has been particularly critiqued within the UNDP, which has developed its own HDI (human development index), notably under the impetus of Amartya Sen's work on poverty, famine and democracy and his critical analysis of the use of purely economic, quantitative indicators, which won him the Nobel Prize in Economics.

5.12

Data on access to drinking water, male and female literacy, the health system and the outcome of campaigns against pandemics, participation in the democratic process, gender and life expectancy, perinatal and infant mortality are all relevant for assessing a society's level of wellbeing and the environmental situation. However, this data is not directly correlated with GDP.

5.13

The UNDPs first composite HDI indicators therefore provoked broad debate and controversy, since the ‘rich’ countries were sometimes well below the top ranking in terms of ‘gross national happiness’. Nevertheless, in view of its solidity (it covers education, life expectancy and incomes adjusted to take account of poverty) this indicator has become the most widely accepted alternative to the purely economic indicators.

5.14

Social statistics are a necessary complement to economic statistics and, in that the public ascribes considerable importance to the key social issues, they also have a political import which it is imperative for governments to take into account.

5.15

Above and beyond narrow economistic thinking and financial short or medium-termism, we must, in all objectivity, recognise that other obstacles stand in the way of this happening, notably, the diversity of social issues and the difficulty of linking and quantifying them with a view to their integration into economic policy orientations.

5.16

We could, intuitively, come to the same conclusions regarding the establishment of environmental indicators aimed at re-integrating external factors into economic growth. For, ultimately, a model of growth that involved the clearing of virgin forest should be rejected, were the impact study to integrate all the social and environmental factors which, as we now know, more than outweigh the economic and monetary factors. On the other hand, it is extremely difficult to put a monetary figure on external factors, such as climate change, the loss of biodiversity, the fate of people who make a living from gathering or using medicinal plants, rapid soil exhaustion and the resulting soil erosion. Although a short term audit might appear largely positive, if these external factors were brought back into the picture, the long-term outlook might well be extremely negative, not only for the region or the country concerned, but for the planet as a whole.

5.17

The objective limits of the use of social and environmental aspects in impact analyses are clearly demonstrated in the cost-benefit approach underpinning a number of the ‘better lawmaking’ assessments and used in EU impact studies (15). Although a relevant indicator (such as number of job losses or lack of opportunities for re-employment) may be used to assess the social impact, the social impact itself is not necessarily a determining factor in the political decision-making process. It often comprises elements that are difficult to quantify, particularly when impact studies cast their evaluations in terms of short-term — or at best, medium-term — monetary considerations. The long term situation is far more difficult to grasp: how does one go about evaluating an aspect such as the expected economic benefits of falling mortality connected with fuel pollution at sea? (16)

5.18

Lastly, debates around social issues use concepts that are sometimes only loosely defined. For example, an indicator on flexicurity would be formulated differently in different countries, or depending on whether it is based on existing experience or a desire to bring the concept into discussions at European (17) or national level with reference to ‘national’ models developed in a particular context and difficult to transpose into other social settings. Which factors should be taken into account and, most importantly, what values, either positive or negative, should be attributed to them? For, ‘the inclusion or exclusion of particular indicators may reveal implicit values and ideologies.’ (18) The problem is magnified still further when it comes to the construction of composite indicators: which indicators should be included, what coefficient should be ascribed to each of them and what is the actual meaning of the composite indicator obtained?

5.19

Composite indicators can integrate both qualitative and quantitative dimensions, can be disaggregated by age, gender and other significant criteria, but must also remain easily comprehensible. To take one example, how could we go about designing an indicator for quality of life in Europe? We could include aspects such as income, life expectancy, perceptions of the efficacy of the health system, pensions, average educational attainment and perceptions of satisfaction with work. But we could also include the unemployment rate, underemployment or housing conditions. And what weight should be accorded to each factor?

5.20

As we can see, the construction of indicators is not a purely technical question. It is closely bound up with a system of shared values or living traditions in a given society. It requires consultation with social organisations and ultimately reflects a particular ideological and political standpoint. At the current time ‘few social indicator approaches actually include societal goals, which are a reflection of social values and standards against which performance may be gauged (…) A crucial element of this approach is the identification and ranking, through consultation and consensus-building, of (…) benchmarks in a number of social areas. Outcomes and inputs, as well as the links between them, are also decided in this process (…). In other words, for social indicators to be policy drivers, process must be part of the product’. (EKOS Inc. Associates 1998: p. 18).

5.21

Another issue which arises is the choice of the subject of statistical analysis: i.e. individuals, communities or the household, as the basic economic and social unit. Likewise, there is the issue of gathering data on particular ethnic groups, which poses problems with regard to the requirement for non-discrimination, but would also be a useful means of determining the nature and scale of discrimination, with a view to proposing policies for reducing and ultimately eradicating it in the longer term.

5.22

The selection of statistics and construction of indicators can be conducted with a view to evaluating an existing policy or with a view to exploring the various options available at an earlier stage. In the latter instance, a broader range of statistical data would probably be needed in order to determine a policy's objectives and the means of attaining them, and could then be narrowed down, once the most relevant statistics and indicators were identified. However, the process of selection is also closely bound up with empirical issues. It is not an exact science and the same statistical data, which include both economic and non-economic data, can be interpreted in many different ways.

5.23

For example, for each of the social indicators used in ‘Society at a Glance: OECD Social Indicators — 2005 Edition’ (see short bibliography), the OECD collected raw data on:

5.24

Eurostat uses the following social indicators:

5.25

To what extent can this (non-exhaustive) list of indicators be successfully integrated into the general objectives of the Open Method of Coordination (OMC) of March 2006:

5.26

Furthermore, the concepts and methods used for certain indicators need to be more precise. In the case of poverty, for example, the French Council for Employment, Income and Social Inclusion (Conseil de l'emploi, des revenus et de la cohésion sociale — CERC) (19) has highlighted the ‘multidimensionality’ of the concept.

5.26.1

Thus, poverty comprises several dimensions: inadequate monetary resources, poor living conditions, inadequate cognitive, social and cultural resources. For each of these dimensions, two approaches are adopted to determine the poverty situation:

5.27

In summary and in conclusion, the social indicators aim to draw the attention of public opinion and decision-makers to social issues which risk being underestimated or poorly understood. Focusing the attention of decision-makers on the key points is particularly important as they generally have to deal with too much information. Herbert Simon put it nicely when he said that too much information kills the information.

5.27.1

From a functional point of view, the purpose of a system of indicators is to achieve ‘optimum aggregation of information’.

5.28

An indicator is more than a statistic:

A system of indicators is not a mere collection of data. It has a number of consequences:

5.29

In practice, the features of an indicator are:

5.30

Features of a parameters system:

6.

The EESC calls for the evaluation of the social impact of EU legislative and policy initiatives to be incorporated in all policy areas. In other words, the Commission should carefully assess the social impact of all relative initiatives without worrying about which DG is in charge of the initiative. This is important if Europe genuinely wants to create a ‘social Europe’ and gain public support. The ‘better lawmaking’ initiative offers a suitable springboard to push ahead in this direction.

6.1

This assessment should examine individually specific groups which might be affected in different ways by the new legislation. Special attention should be paid to disadvantaged groups such as women, disabled persons and ethnic minorities. In some cases, depending on the subject matter of the initiative concerned, it might even be necessary to examine specific sub-groups individually, such as blind persons, for example.

7.1

What emerges from the above and from the public hearing organised by the European Economic and Social Committee on 28 March 2007, is that the multi-dimensionality of certain concepts makes it impossible to develop a social indicator based on a single criterion. By its very nature, the debate on social issues brings into play concepts whose edges are blurred and necessarily vary from one country or one social reality to another. Furthermore, the inclusion or exclusion of particular indicators reveals implicit or explicit ideologies or values. The selection of indicators is closely bound up with empirical issues, and this should, in principle, stand in the way of a rigid approach.

7.2

Although it is laudable and necessary, if not vital, to draw the attention of political decision- makers to the social impact of proposed legislation, there is a methodological problem here, since ‘too much information kills the information’. The European Economic and Social Committee considers that particular emphasis should be placed on the methodology, which has yet to be determined.

7.3

Therefore, the European Economic and Social Committee considers that, at this stage of the reflection process, it is vital to draw the attention of the Commission, and other players, to the quality standards that must be met by all indicators, namely:

7.4

The European Economic and Social Committee also calls on the Commission to incorporate the evaluation of the social impact of EU legislative and policy initiatives in all Community policies, without having to worry about which DG is responsible for deciding whether a social impact analysis needs to be carried out. This is vital if we genuinely want to create a ‘social Europe’ and gain public support.

7.5

The Committee should take full account of the road map and impact assessment alongside the legislative proposal referred to it for an opinion and should begin its work immediately, as soon as the communication accompanying the legislative proposal has been published.

7.6

It is vital to conduct regular assessments of all legislation for which prior impact assessments were conducted and, where necessary, to make adjustments to the way the legislation is implemented and to involve the social partners and, if necessary, stakeholder NGOs in this process. This is important in order to verify the validity of the indicators that were used, to combine them in the social impact assessment and draw lessons or, if need be, persuade the legislator to look into possible changes.

7.7

In certain specific and highly socially sensitive cases (such as labour law), consultations with the social partners should take place at a still earlier stage, so as to identify the most appropriate indicators for conducting as comprehensive and objective an impact analysis as possible.

7.8

There is no doubt that the ‘Better lawmaking’ initiative offers a suitable springboard for pushing ahead in this direction, i.e. proposing vital and effective legislation, whose consequences are both predictable and stable for its addressees and more closely incorporated in the impact analysis and evaluation process conducted by the EU's consultative bodies (the EESC and the CoR) and, depending on the nature of the legislation, by the social partners and the relevant NGOs in the area concerned.