The Ageing Population and Educational Choices

The Finnish Ministry of Employment and the Economy and the Finnish Ministry of Education hosted a Peer Review in Helsinki that brought together ministry officials and independent experts from 10 countries (Austria, Belgium, Cyprus, Czech Republic, Germany, Norway, Serbia, Slovenia, Sweden and the UK), as well as representatives from CEDEFOP and DG Employment, Social Affairs and Equal Opportunities at the European Commission.

“Prediction is very difficult. Particularly about the future.” (Niels Bohr)

This quote appears to encapsulate neatly the one and half days of discussion at the Peer Review meeting. It is not only since the launch of the New Skills for New Jobs initiative by the European Commission in 2008 that there has been an interest in projecting future labour force and skills demands in different sectors and occupations. However, this initiative has certainly galvanised efforts at the EU level to prepare such projections as a contribution to the competitiveness and employment goals of the Lisbon – and latterly the EU2020 strategy.

Finland had previously hosted a Peer Review in 2006 on its long-term forecasting model. Since then, a new foresight process has been introduced which involves co-operation between the Ministry of Employment and the Economy (MEE), the Ministry of Finance, the Ministry of Education and the Ministry of Social Affairs and Health in purchasing future scenarios in the field of labour market, education and training needs. This analysis is produced by the Government Institute for Economic Research (VATT) applying an advanced General Equilibrium Model. At the Peer Review meeting, representatives from MEE and the Ministry of Education presented the benefits of and the assumptions underlying the so-called VATTAGE model. Presentations also focussed on the way in which its projections are being applied in practice to education planning at the national and regional level. Education providers and employer organisations were also present at the meeting and explained their involvement in the process of translating projections into policy formulation. In order to illustrate the pathway from obtaining the projections, over the preparation of different scenarios, to the formulation of policy making, the host country presented the example of the most recent projections regarding the impact of demographic change on education choices.

Finland’s population is ageing more rapidly than many of its EU counterparts. In the absence of significant immigration, dependency ratios are set to increase significantly and demand for health and social care services is set to rise, even when taking into account the higher number of years of healthy life expectancy. The first projection from the new VATTAGE model carried out in spring 2010 confirms that Finland is moving towards a service sector economy and predicts an additional requirement of 120,000 workers in the health and social care sector by 2025. In real terms an increase in labour force requirements on this scale would mean absorbing two entire age cohorts to work in this sector alone. In order to inform policy making, various scenarios have been run involving, for example, a doubling of migration flows (considered to be difficult to achieve in Finland, largely for linguistic reasons), an increase in the effective age of retirement (there has been some success with this in Finland in recent years) and increasing the productivity of workers in the sector. Outcomes in relation to these scenarios show that it is only the latter which would significantly lower the requirement for additional workers in the sector.

The discussions focussed on the various advantages and disadvantages of different models; the implementation of their projections, for example in education planning; and the impact of the demographic imperative on education and wider policy planning.

How to prepare projections and choice of model: In 2009, Finland decided to opt for a General Equilibrium Model (GEM) in the preparation of its forecasts, as – compared to the long-term model previously used – this was considered to be more data rich, more flexible, and to provide the opportunity to generate more reliable sectoral and regional breakdowns as well as enabling the better analysis of different scenarios. Several of the peer countries present – particularly from the other Nordic countries – work with similar models in preparing their projections. Many of the other countries prepare economic, labour market and skills projections with the use of various econometric models, capable of generating demand and supply side projections for a different number of sectors either for the short, medium or long term. On the whole, the national systems used (or under development) were considered to be adequate and appropriate in relation to the level of disaggregated data available in each country. While the Nordic countries were generally regarded to be data rich, representatives from other Member States identified a lack of raw data required to populate an advanced GEM model. Some criticism was made of national statistical offices as well as Eurostat by experts who often found it difficult to extract sufficiently disaggregated and timely data to prepare projections. This is considered to be the result of stringent legislative provisions regarding data protection and privacy.

While all models were considered to provide a useful starting point, none could be seen as sufficient, in themselves, to inform future policy making. In all cases, projections were supplemented with additional quantitative and qualitative data collection and analysis. As well as simple inputs and outputs, models require the inclusion of behavioural and other assumptions which add complexity. The recent crisis is a good example not only to show that economic forecasts have, in this instance, not been particularly helpful in delivering reliable economic, let along labour market projections, but also to demonstrate the need to account and control for such shocks, particularly in short to medium term projections.

How to use the outputs of projections and the choice of scenarios: While some participants went so far as to question the general usefulness of projections and the wisdom of using information from the past to predict the future in an increasing uncertain economic environment, even those who were less sceptical agreed that when it comes to long-term predictions – no matter how desirable they may be – many caveats need to be applied regarding the reliability of all but the most general of this information. In terms of labour force and skills forecasts, many countries therefore limit themselves to short-to medium term projections.

With regard to economic and labour force projections, regional modelling was considered to be particularly valuable, as practical policy horizons in relation to regional development and education provision are often set at this level. Finland is now in the process of using the new VATTAGE model at the regional level and several countries hope to develop similar structures. When it comes to labour force and skills projections, and in particular the need for active labour market policy intervention to address skills imbalances, regional projection have already been used for a number of years by the Public Employment Services in many countries.

As mentioned above, the availability of projections is merely the starting point, providing the “raw material” to inform policy making. To use the example of education planning in the host country, projections are used by the National Board of Education to define some general parameters on the number of education places to be offered in different subject areas. These recommendations then enter the process of “political” negotiations and the agreement of licences with various education providers. In this process, the previous number of places allocated and various vested interests clearly play an important role. Therefore, while projections can serve to delimitate certain overall parameters, this is only the starting point for the political and policy making process. Nevertheless, this process differs from that in many other countries, in particular those (such as Belgium) where individuals are guaranteed an education place in the place and subject of their choice. It was also stated that VATTAGE projections and the model of education planning in Finland was limited to initial education and training and could not be used in planning lifelong learning and the delivery of career change modules, which are increasingly important in closing skills gaps.

How to communicate the model, its underlying assumptions and its results and generate an effective partnership to generate, digest and deal with the policy implications of future projections: The long-term forecasting model previously used in Finland was generally acknowledged to have been highly participative, with involvement from social partners and other stakeholders from the beginning. Other countries placing a high value on social dialogue and social partnership also feature strong collaborative processes in the development of labour market projections. The advantages of such an approach are considered to be the ability to gather more detailed data from a range of stakeholders which can help to calibrate the model or agree on scenarios and assumptions to be included in the model. Such early participation can serve to generate buy-in and increase understanding of the results and their potential policy impact. The disadvantages of such a participative process are considered to be the high cost and time consuming nature of these processes, but also the danger they pose in integrating vested interests into future projections and policy scenarios. VATTAGE was generally acknowledged to be a complex model and in the absence of the early involvement of the social partners (they are now being involved in the discussion of the results), it is not surprising that some criticisms have been levied against the new model. This could be due to the early stage of implementation and much will depend on the method of communicating the model and its results to various stakeholders. While a number of countries argued that they were able to run regional or sectoral workshops to disseminate their results, or indeed – as in the case of Sweden – to distribute their projections to careers guidance officers to inform their advice, it was agreed that more needed to be invested in communicating the methodology, underlying assumptions and results of projections in order to contribute to a more consensual policy process.

The level of co-operation between different ministries in Finland to procure and process projections and scenarios was considered to be exemplary.  A number of countries argued that this would be very difficult to achieve in their own countries, not only because of the existence of policy, but also data silos, which often prove difficult to reconcile.

As well as the need to invest more in the communication of projections and their policy implications, it was also considered important to invest more in the availability of rich, reliable and comparable longitudinal datasets and their analysis from a policy perspective.

How to address labour market consequences of demographic change: One example of the practical policy consequences and implementation of the results of projections was provided in relation to the assessment of demographic trends and their implications of the labour market. As stated above, the first projection of the new VATTAGE model carried out in spring 2010 confirms that Finland is moving towards a service sector economy and predicts an additional requirement of 120,000 workers in the health and social care sector by 2025. As it is clearly not possible or desirable to allow for such a significant share of the workforce to be dedicated to the health and social care sector and thus potentially creating skills shortages for the successful and competitive Finnish high tech sector, different policy approaches need to be developed to address demographic developments and their results. As various scenarios produced by the VATTAGE model show that neither a significant increase in migration, nor a further increase in the effective age of retirement alone can make a significant difference to these projections, a variety of policy approaches is clearly required to activate remaining labour force reserves, postpone retirement, increase the productivity of jobs in the health and social care sector, assist with the training and recruitment of relevant workers in other countries and assisting in the re-training of individuals from declining sectors in the economy towards such high demand occupations. Social partners who are largely responsible for wage setting in these sector also need to be involved in such processes, not only to determine remuneration packages to make such occupations attractive but also to ensure working conditions are such as to support the recruitment and retention of staff.

Participating independent experts

• Robert Arnkil

  Finland
  University of Jyväskylä
• Helmut Mahringer

  Austria
  Austrian Institute of Economic Research
• Magda Kirsch

  Belgium
  Educonsult
• Louis N. Christofides

  Cyprus
  Department of Economics, University of Cyprus
• Daniel Munich

  Czech Republic
  CERGE-EI
• Kurt Vogler-Ludwig

  Germany
  ECONOMIX
• Nils Martin Stolen

  Norway
  Statistics Norway
• Ljubica Spiric

  Serbia
  International Management Group
• Tjaša Redek

  Slovenia
  University of Ljubjana
• Russell Schmieder

  Sweden
  Statistics Sweden
• Bernard Casey

  United Kingdom
  Institute for Employment Research (University of Warwick)