What is a skills forecast and why carry one out? 

A skills forecast is about estimating future labour supply and demand by occupations and qualifications. Different countries or researchers use different names for this exercise, such as employment forecast, labour market projection, employment projection, or labour market forecasting. They all point to the same idea.

The typical time frames for a skills forecast are medium-term (5 to 10 years) or long-term (10 to 20 years). 

A skills forecast is not an ‘exact prediction’, rather it is an attempt to identify the likely outcomes of particular developments and flag up early warnings of possible skills gaps. It is certainly not a ‘crystal ball’ to predict the future! 

Why undertake a skills forecast?

  • To understand present developments and future constraints in order to predict ‘views of a likely future’;
  • To combine complex interactions to provide an easily understandable outcome;
  • To inform various labour market actors on future labour market needs, to help in their individual choices or collective decision-making;
  • To provide an early warning mechanism to highlight any deficits and prevent future labour market imbalances;
  • To support policy-making at the national, regional and sectorial levels in the field of employment, social protection, lifelong learning, career guidance and counselling, and migration

What kind of information should I expect from a skills forecast?

  • In terms of employment, which sectors and occupations will be growing and which shrinking?
  • Which qualifications will more in demand and which less in the future?
  • What will the replacement needs be when certain workers are retired, and how will this compare with labour and skills supply?

Forecast methods

What quantitative methods can I deploy in undertaking a skills forecast?

  • Mechanistic/extrapolative techniques:

Extrapolation is the process of estimating, beyond the original observation range, the value of a variable on the basis of its relationship with another variable. Extrapolation is applied to project, extend or expand known experience into a previously unknown area so as to arrive at an understanding of what may happen under particular circumstances.

  • More complex time series models:

Time series analysis comprises methods for analysing time series data in order to extract meaningful statistics and other characteristics of the data. It is a model that can be used to predict future values based on previously observed factors.

  • Behavioural modelling/econometric models:

Behavioural modelling means using available and relevant data on past behavioural patterns to predict future behaviour. Behavioural modelling is used by financial institutions to estimate the risks associated with providing funds to an individual or business, but is also used in marketing, advertising and sales.

 

Econometric models are one of the tools economists use to forecast future developments in the economy. In the simplest terms, econometricians measure past relationships among such variables as consumer spending, household income, tax rates, interest rates, employment, and the like, and then try to forecast how changes in some variables will affect the future course of others.

  • Skills audits:

A skills audit is a process aimed at identifying and analysing the knowledge, skills and competences of an individual, including his or her aptitudes and motivations, in order to define a career trajectory and/or plan a professional reorientation or training project.

What qualitative methods can I use?

Qualitative methods are also sometimes used in skills forecasting, such as Delphi techniques (expert opinion/consultation), case studies, focus groups, holistic modelling approaches, scenario planning (plausible futures), etc.

The Delphi method is a structured communication technique, originally developed as a systematic, interactive forecasting method, which relies on a panel of experts. The experts answer questions in two or more rounds. After each round, a facilitator provides an anonymous summary of the experts’ forecasts from the previous round as well as the reasons they gave for their judgements. Thus, participants are encouraged to revise their earlier answers in light of the replies of other members of the panel. It is believed that during this process the range of responses will decrease and the group will converge towards the 'correct' answer.

 

The Holistic modelling approach is concerned with looking at whole or complete systems rather than analysis based on dissecting them into parts.

Generally, however, quantitative methods are deployed only as complementary tools, and the main methods remain quantitative.

What are the most widely used forecasting techniques? 

  • Quantitative modelling: combining behavioural/econometric models and extrapolative techniques:
    • Multi-sectoral, dynamic macroeconomic models (MDM) or
    • Computerised general equilibrium models (CGE);
  • Projections of the occupational structure of employment within each sector (increase, decrease, replacement);
  • Demographic trends and projections, including changing patterns of qualifications.

What data sources are needed to carry out a skills forecast?

  • Economic datasets: national accounts, input-output tables, growth patterns data, exchange rates, interest rates
  • Labour force survey (LFS) datasets
  • Education and training statistics

► Long-term time series data on employment by sector, region, occupation and qualification level

Other potentially useful data:

  • Population censuses, other household surveys
  • Establishment censuses, enterprise surveys/statistics

► The quality of the data is key since the forecast results cannot be more robust than the data upon which they are based.

Steps in undertaking a quantitative forecast:

1. Developing a ‘mathematical model’ which reduces the complexity of the reality into a predictable future.

► The more complex the model is, the closer it is to reality, but the more difficult it is to manage and use in forecasting future scenarios. 

2. Collecting long-term time series data concerning past trends in the economy and labour market, as well as with regard to demography, education, etc.

► Forecasting is a resource-intensive process, requiring substantial prior investment in regular data production and analyses. The existence of regular and reliable data (e.g. national accounts, population censuses, labour force surveys, establishment censuses) and electronic access to these datasets are vital.

3. Making assumptions on the future of the economy and labour market as well as in terms of demography, education, etc. 

► The key assumption of skills forecasting is that the patterns of performance and behaviour in the economy and labour market will reflect past trends and that there will be no major disruptions to the economy.

4. Choosing and developing your own initial forecast modelling as a 'pilot version'.

► This modelling should not be viewed as 'production ready', and should be used for trial purposes and for identifying and discussing initial outcomes.

5. Repeating subsequent rounds of forecasting (e.g. every two years) to better understand the process and the interpretation of the results

► Evaluating the usefulness of the outcome is crucial, NOT how well the model predicts exact numbers!

6. Adding new elements into the forecast model in every round to make it more complex but also more realistic.

► You need time for consolidation and to establish the excellence of your forecasting model!

What are the key elements of a skills forecast?

In general, a full forecasting exercise includes five elements or so-called ‘modules’:

1. Macro-economic: Multi-sectoral, dynamic projections;

2. Employment change: Employment estimates by sector, occupation, qualification – in terms of expansion or contraction;

3. Replacement needs: Permanent/semi-permanent withdrawals from the labour market;

4. Labour/skills supply: Labour stocks, flows, education choices;

5. Imbalances: Estimation of demand and supply mismatches.

Challenges and problems faced in skills forecasting 

What could go wrong in my forecasting exercise?

  • Data problems can cause misleading results, e.g. lack of or incomplete datasets, the quality and comparability of data issues, lack of long time series, etc.

► Gathering and harmonising all the relevant datasets requires extra work due to frequent changes in the classifications and definitions used. The lack of good quality, reliable, regularly updated and detailed data on occupational structures within sectors is often a problem. It might be necessary to first develop a long time series on employment by industry (economic sector), status, region and occupation.

  • Wrong and/or changing assumptions could lead to overly optimistic or pessimistic scenarios.

► Predictions will deviate from reality if the assumptions behind them are flawed. The reliability of forecast results depends on the concept that the past is a good predictor of future developments.

  • Unexpected changes and/or radical disruptions can change the course of history.

► The financial crisis of 2008 was a particularly dramatic deviation from the near past. In such cases the forecasting teams' skills in adjusting and interpreting past developments to fit likely future scenarios are crucial.

  • What kind of obstacles might I face regarding skills forecasting?

Typical obstacles in developing a skills forecast:

Insufficient funding, lack of technical capacity, data quality issues, coordination problems, lack of interest and the low reliability of results can all pose major challenges for skills forecasting.

  • Typical obstacles in using skills forecast results:

► Methodological issues: results are too general as a result of a lack of occupation or sector-specific data; or there are flaws in the way skills are measured.

► Limited stakeholders’ knowledge and involvement in the exercise as well as in the discussion of the results. 

►Scattered responsibilities in responding to identified skills needs, lack of coordination and leadership for taking action.

Who should be involved in a forecasting exercise?

  • Independent research institutes or university centres, specialising in labour economics;
  • State statistical offices;
  • Ministries of economy or labour;
  • Ministries of education or VET;
  • Social partners;
  • Semi-public or private agencies, councils or commissions.

Be the first one to comment

Please log in or sign up to comment.