EU STRUCTURAL POLICY INDICATORS FOR THE IDENTIFICATION OF PROBLEM REGIONS
|3.2. Structural Fund Indicators
|3.3. Recommended Indicators - Immediate|
|3.4. Suggested Indicators Longer Term|
This chapter analyses the current indicators for Structural Fund eligibility. It then suggests some alternative indicators which might be used for the identification of problem regions. This is divided into two sections: indicators which could be available in the short-term and indicators that could only become available in the longer term following changes in either collection or mindset.
3.2. Structural Fund Indicators
The current (and proposed) criteria for the vast proportion of Structural Fund spending through Objective 1 is clear cut. With the qualifying criteria set at NUTS II regions of less than 75% of EU average GDP, the allocation goes to those regions with the lowest GDP.
But the same regions would not necessarily qualify if faced with the Objective 2 test of unemployment. An analysis of unemployment and GDP for 180 NUTS II regions finds that roughly half of the regions with below average unemployment also have below average GDP (7). An earlier study commented on the relationship between unemployment and GDP at the regional level that,
"there is little correlation at regional level between GDP per head and unemployment. In particular, Greek and Portuguese regions combine some of the lowest per capita GDP levels in the EU with some of the lowest unemployment rates. There are many reasons for this lack of correlation, including differences in labour market organisation, varying levels of productivity and underemployment, and variations in activity rates. Spain and Ireland in particular stand out as having both high unemployment and low activity rates; in both these countries, GDP per head is depressed by the fact that low proportions of the population are in employment. In contrast, in Greece and Portugal per capita GDP is low despite high employment levels, because labour productivity is exceptionally low." (8)
Objective 1 and Objective 2 criteria measure different things. Objective 2 is not a scaled down version of Objective 1. In a recent speech on the issue the Commissioner stated that "I believe that the reduction of disparities between regions and measures accompanying structural change address two different problems requiring differentiated solutions." (9)
This clear statement then requires two strands of questioning. Is this the correct understanding of what the Structural Funds should set out to achieve? And, if so, have the correct indicators been chosen?
It is not the place of this report to consider at length the former question. But we would note in passing that such a distinction also implies the question as to what is the appropriate balance, in terms of allocation of resources, between the two problems.
Rather we take these as the given policy goals and examine how improving the indicators can help ensure the Structural Funds are best targeted to achieving these goals, consistent with other EU objectives.
We set out in the next two sections some suggested indicators for ERDF interventions. We have distinguished between those which are readily available at present and could be introduced in the short-term and indicators which would require changes in collection and thus could only be used in the medium-longer term.
3.3. Recommended Indicators - Immediate
For the short-term we suggest looking at a number of indicators which might be considered complementary to those which are currently used. This will not produce any radically different effect but should achieve better targeting of the underlying objectives by providing a more comprehensive profile of the regions. For these immediate indicators we have been guided in particular by datasets held on Eurostat databases. As in the previous chapter's review of the Structural Funds, we discuss in turn indicators related to Objective 1, Objective 2, and the urban agenda of Objective 2.
We have discussed in section 2.5 above the problems associated with GDP as a measure, yet this still remains at present the most acceptable measure of the productive wealth of an area. Objective 1 relies on spatial targeting of large areas and assume productive wealth is the appropriate measure. In this regard some complementary measure of the distribution of income would appear appropriate to ensure that the benefits of the increased productivity capacity of an area are widely enjoyed.
Income Distribution expressed as a Gini coefficient this measures the distribution of income for a given area. Gini coefficients are not readily published at NUTS II level, but estimates of income distribution are available. An alternative would be to express the indicator as a ratio of the income of the top 10% of earners to the income of the bottom 10% of earners.
But if household income is the appropriate measure then it would require different eligibility criteria. In this case it would not be the productive wealth of an area that is measured but the income of people. A further complementary indicator to assess whether providing aid to the poorest regions also targets the poorest people would be the percentage (or number) of households living below the poverty line.
Percentage Living Below the Poverty Line The EU has a standard definition of poverty, which is those on less than one half average income. Estimates of income data are widely collected on a regular basis and whilst not readily available at present there should be no practical difficult in compiling this indicator.
In contrast to Objective 1, the indicators for Objective 2 are employment and unemployment. Unemployment enables good spatial targeting as it is widely available for small local areas. But unemployment does not measure the full extent to which there is under-use of labour, and even with standardised definitions of unemployment, percentage rates may not be strictly comparable from one member state to another. Rates of economic activity are a good complementary indicator expressing the under-use of labour which the EU has identified as a problem. Analysis of UK figures shows a higher number of people enter employment from the economically inactive list than from the unemployed register.
The labour market can also be analysed in terms of employment. Yet employment does not represent an inverse measure of unemployment. The relationship is a lot more complex. An article from the UK Labour Market Trends described the importance of analysing labour market flows if the true position was to be assessed:
"Unemployment figures only provide a partial picture of the labour market. A much more useful picture is provided by complementing these with figures for employment and economically inactive people. Net changes in, say, employment arise from a pattern of flows in and out of the other two labour market states. To interpret an increase or decrease in unemployment, compatible figures are needed for employment and economic inactivity as well. For example, in some situations an increase in unemployment may be accompanied by an increase in employment, whereas in others it will be accompanied by a decrease. There are important differences in how these two situations should be interpreted. A slowing in the rate of employment increase may be as important an indicator as an increase in unemployment. Hence employment data are important in making an assessment of labour market conditions. Looking at an unemployment figure in isolation gives an incomplete picture which at times may be misleading." (10)
Therefore in order to provide a more accurate profile of the state of the labour market it is necessary to look at indicators of
Employment - disaggregated by sex
Unemployment - disaggregated by sex
Economic activity - disaggregated by sex
But all labour market indicators can be volatile. Because of the relative flexibility of labour employment and unemployment vary significantly with the economic cycle. When using these indicators to identify problem regions it is important to distinguish the structural from the cyclical measure. Figure 3.1 shows the change in unemployment rates over the course of the economic cycle for the UK and how this compares with output and employment measures.
The current three year period for measuring unemployment is not sufficiently long to be independent of cyclical effects. As such the eligibility of regions can be dependent on the relative position of member states in the economic cycle. A five year average, caveated by excluding those regions on a continuous year on year improvement, would account for the worst excesses of such cyclical effects without affecting the structural objectives of the criteria.
This could be further reinforced by including data on long-term unemployed as a complementary indicator. There is a strong correlation between long-term unemployment and structural unemployment as workers whose skills have been made redundant have difficulty being re-integrated into the labour force as the pool of industrial jobs shrinks.
The structural problem is also a reflection of the lack of technological innovation. Perhaps as a monitoring rather than eligibility indicator it would also be worth looking at a measure of technological innovation.
New Patents Issued - this data is maintained by Eurostat at NUTS II level and is more easily collected than alternatives such as % of R&D expenditure.
The Urban Agenda
There is a specific urban agenda to address. Densely populated urban areas have higher rates of unemployment and higher rates of long term unemployment. The communication, 'Towards an urban agenda in the European Union' notes "In cities, multiple deprivation is expressed in rising poverty and homelessness, by social isolation, bad housing conditions, drug abuse, and criminal behaviour" (11).
The recognition comes in the proposed new Objective 2 criteria.The proposals allow for considerable discretion in defining indicators of urban poverty. There are advantages in that this recognises the distinctive characters and problems of different areas. It does not help in enabling comparability between regions when determining eligibility. The new criteria imply indicators that are able to measure at NUTS III: long-term unemployment; poverty; degraded environment; crime; low educational attainment. We suggest a readily available measure for each of these below:
Long Term Unemployment - Long Term Unemployed measured as the percentage of the total unemployed who are long-term unemployed (those that have been unemployed for more than 12 months). This is the hardest category to reintegrate into employment and a major indicator of social exclusion. Data is available at NUTS II.
Poverty - Percentage Living Below the Poverty Line. It is a reflection of people's ability to participate in the economy and society. The standard EU definition of poverty, is those on less than one half average income. Estimates of income data are widely collected on a regular basis and whilst not readily available at present there should be no practical difficult in compiling this indicator.
Degraded environment - Derelict Land and Buildings. This data is not immediately available, but could be quickly compiled if there were a priority to do so. It would be measured as i) the percentage of the total area which is contaminated, degraded and derelict, and ii) the percentage of the total building stock which is vacant or derelict.
Crime - Number of burglaries or violent crimes per 1000. The purpose here is to identify those crimes which residents may have an expectation of experiencing. The fear of crime is a major factor in blighting urban living for significant proportions of the population.
Education - Percentage of 16 and 17-Year Olds in Full-Time Education. It is difficult to establish comparable measures of qualifications across member states. But the quantity of education as represented by the proportion who remain in education beyond a certain age is readily available and measurable. This will be expressed as the percentage of those of ages 16 and 17 who are in full-time education.
If the special problems facing Island regions are to be addressed by the Structural Funds, then indicators which reflect their comparative disadvantage need to be introduced. There is a strong correlation between levels of GDP and peripherality as emphasised recently in the document European Spatial Development Perspective (12). Drawing on the discussion in Chapter 2 there are three types of indicator which reflect the disadvantages suffered by Island regions. We suggest examples which might be used although further work is required to develop the most appropriate indicators.
Accessibility Index - can be represented by standardised costs or travel times. Such indicators are used in the freight and transport sectors and can be developed from this source. For example, an indicator of the accessible population within 8 hours travelling time was illustrated in 'Europe 2000+' (13). This shows the scale of markets that can be reached from a particular region.
Infrastructure Endowments - indicators such as km of rail, road or bus services are held on Eurostat databases. But in this context indicators of utility capacity are most relevant. For example, the capacity to supply adequate amounts of clean water and to dispose of solid and liquid wastes generated.
Social Infrastructure - might be represented by travel time to nearest comprehensive healthcare facility, or higher education institution.
When using these indicators however, it is necessary to distinguish between interventions which are intended to improve the economic structure and competitiveness of the region and transfer payments which are intended to compensate the region for its disadvantage. The latter two indicators will carry strong elements of transfer payment which may not be appropriate to the Structural Funds.
Member State Priorities
An alternative approach might be to recognise the differences that exist in data collection and usage at member state level and to fully embrace the concept of subsidiarity even with regard to allocations. This will enable member states to prioritise structural issues and areas of greatest concern, and hence use and develop indicators which best reflect these priorities. But such an approach would necessitate a weighting system which could account for these differences to enable transparent allocations of funding.
An advantage of this approach, however, is that it enables the full richness of data sources to be exploited. For example, in making the case for the Urban strand, each member state would be able to draw upon its own data on income distribution, households in receipt of welfare benefits, health statistics, educational attainment, housing conditions, and other data.
It would also be able to use locally developed deprivation indices, which would at least be internally consistent to the member state. The UK has developed an 'Index of Local Conditions', a composite indicator which, for the local authority districts of England, weights 13 factors: the level of unemployment; the ratio of long-term to all unemployed; Income Support recipients; children in low earning households; households without a car; standardised mortality rates; low educational (GCSE) achievements; participation in education of 17 year olds; house contents insurance premiums (as a proxy for crime); children in unsuitable accommodation; overcrowded housing; housing lacking basic amenities; and derelict land.
3.4. Suggested Indicators Longer Term
For the longer term we can consider both expanding the selection of datasets and also a different conceptual approach. There are four particular areas we examine: taking on board themes such as sustainable development; modifying existing criteria; introducing new datasets; and revised forms of GDP.
In an earlier report, Sustainable Development: A Key Principle for Economic Development', we set out a hierarchy of indicators covering a range of economic, social, environmental and institutional factors which we believed could be used to monitor sustainable development. The top tier, or headline indicators of this hierarchy are set out in table 3.1 below.
This helps operationalise Sustainable Development as a theme and sets a framework which can, as noted by the Sustainable Cities Report of 1996, 'promote an integrated approach to urban problems encompassing social, economic, and environmental factors'. (14)
Table 3.1. : Headline Indicatorsfor Sustainable Development
GDP per Capita
Energy Consumption per Capita
16-17 year olds in Full-Time Education
Land Use and the Quality of the Urban and Natural Environment
Waste Generation and Management
Qualitative Assessment from Secondary Indicators
Qualitative Assessment from Secondary Indicators
Qualitative Assessment from Secondary Indicators
Qualitative Assessment from Secondary Indicators
Qualitative Assessment from Secondary Indicators
For the longer term there is the possibility that the Objective 2 criteria may need to be revised away from its dependence on industrial employment. Other forms of structural dependence and weakness are likely to emerge with the long term change in the sectoral balance of employment in favour of the service sector. A new typology may need to be developed which distinguishes whether sectors are in long-term decline, rather than their classification of industrial or service.
For the longer term some measure of the quality of employment would be a valuable addition to the monitoring of structural weakness. Whilst this is an objective which could be considered in the short term, the practical technical difficulties of agreeing an appropriate measure militate against this. One possible indicator is the proportion of part time employment but this is ambiguous as a high proportion of female part-time employment is through exercising preferred choice. There are a number of types of measure which might be used for quality of employment:
Infrastructure endowments are recognised as being one of the principal determining factors in regional development performance and a large proportion of Structural Fund expenditure has been used for infrastructure projects. Yet infrastructure indicators do not constitute explicit qualifying or allocation criteria. There is an acknowledged difficulty in developing an appropriate indicator. As the EC report 'Competitiveness and Cohesion: trends in the regions noted, "For many peripheral and lagging regions, a key problem is the deficiency of the internal network rather than the inter-regional links, which is not always revealed by any regional indicator, since this will not take account of how well different parts of the region are connected." (15)
The accession of countries from Eastern Europe for the post 2006 Structural Fund round will have profound implications. Nearly all regions of the candidate countries would qualify for current Objective 1 status, even following the support provided in the Instrument of pre-Accession. This will have two significant effects. It will run counter to attempts at concentration of resources by spreading the number of eligible regions. But at the same time the mathematics will mean that an influx of regions with below EU average GDP will lower the average GDP. Hence existing Objective 1 regions may find that although their absolute position may not have improved, their relative GDP to the new lower average has risen above 75%. Either the simple criteria will need to be amended or parallel forms of assistance for new and old member states will have to be established.
Data for new indicators is being gathered, especially in the field of environmental indicators. These indicators will be used for evaluation and impact assessment but can also be used to identify, for example, the scale of environmental degradation of urban areas. Data on environmental quality is increasingly available at the regional (NUTSII) level as a consequence of a number of parallel initiatives including: the monitoring and reporting activities of bodies such as the OECD, Eurostat and the European Environment Agency; the monitoring requirements of various EU directives (for example on water quality and on air quality); and the higher priority being awarded to environment at the local authority level as a consequence of Agenda 21 and various related EU initiatives.
At present there are variations both in data availability and consistency in many areas. Recent initiatives, for example from Eurostat and the OECD, have sought to improve data availability and to harmonise the approaches used to collect and analyse data. In particular we would look to the development of standard indicators for: Air Quality; Water Quality; Land Use and the Quality of the Urban Environment; and Waste Generation and Management.
We would also expect to be able to make better use of health and other social data which can better define the disadvantage an area experiences.
New Measures of Welfare
We discussed in section 2.5 above some of the limitations with GDP as a measure of welfare and some of the initiatives that have been undertaken to try and rectify this. Development of new indicators in this area are only likely to function as part of a very long term project, due to the 'acceptability' test discussed earlier. We would restate the comment made earlier that work in this area should continue, but a system of parallel indicators integrated into the standard assessment is a more productive way forward.
In the short-term any modifications which may be made to Structural Fund criteria will be limited. Without amending the principal criteria we would suggest that subsidiary indicators can be introduced in order to better target Funds to their underlying objectives. These could be used to help determine allocations within the qualifying regions.
Modifications may also be made to existing criteria to better eliminate cyclical effects, specifically with regard to the unemployment qualifying period for Objective 2.
In the longer term there will need to be more fundamental reform of the Structural Funds as new poorer member states from Eastern Europe join the EU. A longer term review will also provide the opportunity to look at introducing more complex eligibility and monitoring criteria which integrate environmental and sustainable development themes into the decision making process.
7. Competitiveness and Cohesion Trends in the Regions' EC , Statistical Annex Unemployment rate average 91-92-93, GDP per capita PPS average (89-90-91), 1994.
8. The Regional Impact of Community Policies, 1996, Roger Tym & Partners for European Parliament. See also Dignan, T. 'Regional Policies and Regional Disparities in the European Union', in Oxford Review of Economic Policy, Summer 1995.
9. Wulf-Mathies, M., The Lessons of the Past, Pathways to the Future.
10. Labour Market Trends, March 1996.
11. Communication from the Commission Towards and Urban Agenda in the European Union, Nov. 1997.
12. European Spatial Development Perspective, EC, 1997.
13. Europe 2000+, EC, 1994.
14. Sustainable Cities, European Commission, 1996.
15. Competitiveness and Cohesion: trends in the regions, Fifth Periodic Report, EC, 1994.
|© European Parliament: September 1998|