So, le tout Ottawa now seems convinced, given that a) the March budget is allegedly about skills (for the middle class, you know), b) the feds mostly handed the skills portfolio over to the provinces years ago that Individual Learning Accounts (ILAs), are definitely On The Agenda.  Possibly with some language around guaranteeing workers time off for skills training.

So, can this work?  Has it worked elsewhere?  Glad you asked.

The idea of ILAs are nothing new.  In one form or another they’ve been around for about thirty years and have been tried in over a dozen countries.  But their implementation has varied enormously and details matter, so it’s hard to draw general conclusions.

The basic idea is that individuals have special accounts to which they may contribute, but more importantly governments can periodically dump money into them for a specified purpose (they can be considered policy-adjacent to vouchers).  Sometimes growth in the account is tax-exempt, sometimes there are various matching schemes, etc: the scope for minor policy variation is pretty much endless here.

But there are two major policy questions which need to be settled at the outset of any program.  The first is: does everyone get an account? Some systems deliberately restrict access to these accounts to particular groups of people – normally some variation on individuals with low-income or with low-skills or occasionally (as in Germany under the Hartz concept) job-seekers.  Obviously the cost and distributional impact of the system can be tightly controlled under these kinds of systems.

The second question is: if everyone gets an account, are the top-ups equal across the board?  Obviously, this is a much more expensive type of program, and a more regressive one, if you think about it.  Not everyone who receives money for retraining uses it, and those likeliest to take training are those who already have high levels of education (i.e. the wealthiest).  For the same reason, the universal approach is the policy option with the highest prospect for “windfall gains” to taxpayers, who would have undertaken training even in the absence of the subsidy.

There is a third piece of the puzzle, which is not as well covered by the international literature because in most countries it’s not a consideration: what counts as eligible training for the program?  This always matters because the kinds of training that are envisaged as being supported by this kind of subsidy (rarely worth more than a few hundred dollars per year) are predominantly short-course topics not delivered by post-secondary institutes governed by complicated quality assurance mechanisms.  In places like France or Singapore (currently the two “hot” countries as far as experiments in ILA policies go, and two examples I’ll delve into in future blogs), governments have a pretty good handle on who the trainers are and can issue some pretty minutely-detailed instructions about what kinds of courses do and do not qualify under the account system. 

Canada, to put it mildly, does not have any infrastructure for this.  We do not have a national database of trainers.  We do not really have a microcredential system that would encourage provision of this kind of training.  We have literally none of the infrastructure that would make this part of an ILA system easy to implement (I recommend everyone read the UK National Audit Office’s report on that country’s smashing failure of an experiment with ILA’s in the early 2000s, just to get an idea of how badly this idea can go wrong). 

Of course, money isn’t the only barrier to adult education.  There’s also the issue of time, which is why it’s possible the Liberals are thinking about creating training time entitlement in addition to ILAs.  These exist elsewhere (notably France) and within Canada the idea was been kicking around for at least 35 years; the MacDonald Commission (1982-85) did some intriguing studies on something it called a “Time Bank”, which (if I recall correctly) would effectively have acted as an EI-based entitlement to time off (3 days a year, for instance, to be carried forward if needed).  This is a little bit trickier to arrange because a) business will scream and b) not all training opportunities occur during working hours (how do you make it fair to people who want to take night classes?) but it’s not an insurmountable problem.

What’s the takeaway from all this?  ILAs for all are probably a good idea, but equal subsidies for all are not.  In fact, at the upper end of the income distribution the subsidy should be as close to zero as possible (let people save in a TFSA-like account if they like, but no top-ups) and at the bottom end should be quite generous (over $1000 per year).  Guaranteed time off for training is a great idea and would probably even more transformational than the money.

Make no mistake, the practical problems governing eligibility criteria for use of the money are huge and potentially fatal.  It would be…courageous…to launch a signature program without having all the details worked on this score.  Since I highly doubt that the details have been worked out I hope that whatever gets announced at least has a long pilot phase built in (perhaps limiting use of the accounts to a single popular field like project management or certain types of IT, where it would be possible to more easily monitor courses and providers).

But I guess we’ll see next month.