Lifelong learning – the evidence and the responses

I was invited to give a presentation on life-long learning to the APEC Future Education Forum held (virtually) in Seoul on 30 September 2021.  The forum’s theme was: Strengthening lifelong competencies and skills development for individuals’ career, education, training and life cycles.

Below are the slides and speaking notes I used in the presentation.

Thank you very much for the opportunity to participate in this conference – the theme of the conference is a very important one for all countries.  Countries in our region are no exception.

Lifelong learning (LLL) – or recurrent education for adults in the workforce – is (or should be) one of the most important areas of interest for education and labour market policy makers.

Yet it’s an area where we have relatively less data on which to make policy decisions.  We have great data on formal, credentialed post-secondary education because that sort of education usually attracts funding.  Funding attracts attention.  And it’s also part of young people’s life course – it’s inherently interesting.

But much LLL is informal and non-formal and unfunded.  We know less about it.

In the last few years, there has been a great advance – the OECD’s Programme for the International Assessment of Adult Competencies (PIAAC) includes the Survey of Adult Skills (SOAS) which tests respondents’ skills and asks about LLL alongside background questions about education, employment, social background …. That has revealed great data on LLL which sheds light on the policy issues.

My presentation looks at what we know about LLL – from PIAAC and other sources and asks what that tells us about the policy choices we face.

We need to put this in context.

Our countries face changing demography.  Nearly all developed countries have falling birth rates and falling fertility rates. And extended life expectancy.  We are getting older.

As an extreme example case study …. Look at Japan.

By 2050, Japan’s population will be around 20 million less than now.  The proportion of young people is falling steadily.  And the Japanese have the longest life expectancy in the OECD so the proportion of older people is set to become much greater.

That has major implications for the labour market – fewer people to do the work and even fewer coming through to enter the workforce.  And a growing number of older people to support. 

Let’s put that alongside the picture of demand for skills.

This data is a bit old but what it shows is that firms in many countries face difficulty recruiting the staff they need.  All developed countries are facing this problem, especially as the educational attainment of their population rises, meaning greater mismatches in the labour market.

Japan, again, stands out.

My own country, New Zealand, is way to the left on that scale also – fifth with 50% of companies unable to fill vacancies.  This is a problem that New Zealand has usually addressed through an active skills-based immigration policy – a policy that carries considerable cost for national infrastructure and for the dynamics of the labour market …. leading to political cost for government.

And, of course, we all know about the 4IR, the growth of automation and its effects on employment….

Is this the magic bullet that mitigates the demographic challenge? Is that how we deal with the shrinkage of the working aged population?

In short … no.

The original claims about how robotics would take nearly all of our jobs was based on 2013 research by Frey and Osborne which was based on job titles. Frey and Osborne painted a picture of an apocalypse.

But one of the spin-offs from the PIAAC programme is that analysts can look at the actual tasks people perform at work and assess which tasks (rather than which jobs) will be able to be automated.  This graph shows that a high proportion of jobs face change, around one in four face significant change – most being transformed, rather than disappearing.

Changes resulting from technological change will mostly mean that most jobs will require more and increasing skills.  Increasing skills will mean greater demand for LLL.

So what does PIAAC tell us about the take-up of LLL?

Around 40% of workers had undertaken some form of job-related training in the 12 months before the survey.  It’s higher in the Nordic countries, my country New Zealand, and also, among APEC countries, it’s high in Australia, the US, Singapore and Canada.

And how much training do workers do? 

In this graph, the bars show non-formal training – that is organised classes but not leading to a qualification.  The diamonds show informal – essentially self-directed learning. 

The average is about 45 minutes of non-formal training a week (non-credentialed training) and 1 hour of informal training (self-study).

The SOAS asks about people’s attitudes to training – whether they are wanting to do more training, whether they consider there is value in training, how they get information, how they deal with new ideas.  From those responses, the OECD constructs a Readiness to Learn index.

Four APEC countries appear in the left-hand group in this graph (US, Chile, Canada and New Zealand) and two on the far right (Japan and Korea).  This raises an important set of questions about “why don’t some workers take more training?”

Among those motivated to take more training, the greatest reported barriers relate to time and busy-ness – busy-ness at work or home, the black and grey/blue blocks.

Cost is an issue, but not as large an obstacle as time.

What may also be a concern (but doesn’t emerge as a reason there in this OECD analysis) is that there may be doubts about the quality of extra training and, possibly, the rewards from training.

How do governments respond to the need for training and also the evidence for barriers?

The traditional response has been that this fringe area of the education system gets overlooked in the contest for government’s attention.  That needs to change and is slowly beginning to change, as the kind of skills needed in the workplace change and as shortages of skills begin to bite.

Uncertain quality and the lack of recognition of training are important.  If people have anxieties about their ability to fit training in to busy schedules, they won’t want to take a full quality-assured credentialed qualification.  Neither non-formal, nor informal training is visible on a CV.  And besides, the evidence for good labour market returns for extra full qualifications undertaken by qualified workers is very mixed.

Micro-credentials are bite-sized, they give recognition, they can be quality-assured …. They can be designed so that several micro-credentials can be stacked/grouped to make a component of a full credential.  It’s great to see the New Zealand government has moved to formalise and quality-assure micro-credentials and that some training providers are offering them now.

How much should the New Zealand government pay for or fund LLL given the high take-up?  There is some funding in that some workers/employers use the credentialed system for their upskilling training – such as the MBA and similar qualifications. And employers usually pay for their staff to do some training if it is essential for the work and closely linked to the requirements for the job.

So that raises the question about dead weight – would government end up paying people to do no more than they were already planning to do at their own expense?  Or would funding encourage greater take-up?

Of course, the counter-factual is not easily knowable – we don’t know if and how much the take-up of training would rise in the presence of funding unless we do it.  And once you do fund something it’s awfully hard to take it away if it does prove to be deadweight.

In Singapore and Korea, the governments have introduced a lifetime learning account – a training voucher.  That’s probably an interesting thing to try.   And in New Zealand, the government has placed a toe in the water, introducing funding for short learning packages targeted at people who are aged 25 or over and who are affected by redundancies and technology disruptions, or upskilling because of the technological demands of existing roles.

But the main barrier to LLL is time.  If you take a day out to attend a course, does the work just continue to pile up, so you bear the cost in stress as you try to catch up?  Do you end up having to do preparation in the evening when you should be reading to your children?  That’s something not in a government’s control.  The role of government should be to encourage employers and to broker or facilitate.  Perhaps paying more towards LLL might be part of the encouragement that they give to employers to enable workers to have time to train.

LLL – such an important issue, one that is becoming increasingly urgent. Our governments need to be thinking about and working on this.

The whole forum is available for view on YouTube. My slot is nearly two hours in …. 1:50.46.


Frey B and M Osborne (2013). The future of employment. How susceptible are jobs to computerisation?

Manpower Group (2015), Talent shortage survey

Nedelkoska L and G Quintini (2018), Automation, skills use and training

New Zealand Productivity Commission (2020) Technological change and the future of work

New Zealand Qualifications Authority (2018) Micro-credentials in New Zealand’s education and training system: a consultation paper

New Zealand Qualifications Authority (2018) Micro-credentials system launched

OECD (2021) Micro-credential innovations in higher education: who, what and why?

OECD (2021) Quality and value of micro-credentials in higher education: Preparing for the future

OECD (2021) Skills outlook 2021

OECD (2017) Skills outlook 2017

OECD (2017) Labour force statistics: population projections

Smyth R (2020) The future of work – and what it means for tertiary education

Smyth R (2018) We need to talk about … life-long learning

Tertiary Education Commission Guidelines on increasing SAC3+ funded delivery of short learning packages

Wheelahan L and G Moodie (2021) Analysing micro-credentials in higher education: a Bernsteinian analysis

World Bank (nd) World Bank data: fertility rate and birth rate