In the past few weeks, I’ve chaired events and held discussions at a number of events looking at how local government could build better learning, skills and employment systems.

I’ve come out of them with a growing sense of optimism about the opportunity ahead. Optimism is in relatively short supply in many places at the moment, so let me explain why.

There are a lot of challenges involved with the current approach to devolution.

It’s partial, focusing only on some of the adult-education budget in some areas; it’s complex, with a range of issues such as how to make sure the system works smoothly for people whose nearest college or provider is in a different geographic area; and it comes amid a context of sharp cuts to local authority funding and an uncertain spending review ahead.

‘That’s your problem, it’s devolved now’

We need to avoid what in the US would be called an unfunded mandate - the transfer of responsibility, but without sufficient funding to carry out that responsibility. In the years ahead, if the mayors of combined authorities complain that they don’t have enough funding to meet skill needs, the answer from government cannot be “that’s your problem, it’s devolved now”.

These are all very real challenges. But there’s a prize to be gained. And that is to build a more effective employment and skills system that works for people and employers.

Some critics of devolution worry about a postcode lottery. But we have a postcode lottery for outcomes now - for example, the Learning and Work Institute’s Youth Opportunity Index shows how much education and employment outcomes vary for young people across the country.

In my view, tailoring services to local circumstances offers the opportunity of levelling up outcomes, rather than levelling across services.

‘Joined-up approach needed’

To make this work, though, we need a more joined-up approach to devolution than we have currently. For example, our work with the Local Government Association has highlighted Canada’s labour market development agreements as a potential model to follow. These set out the range of services and funding the federal government will devolve to the provinces and the outcomes - such as sustained employment - which the provinces commit to achieving.

Meet these goals and more devolution will follow. Fall short and the Canadian government will intervene.

We need our mayors to set out how they want to do things differently, given more powers. We’re getting some signs of this with, for example, Sadiq Khan saying he wants to tackle in-work poverty by extending free training for those earning less than the London Living Wage.

A skills congestion charge?

What other innovations will local areas want to make? The early days of the London mayoralty featured the introduction of the congestion charge and the Oyster card.

What are the equivalent step-change innovations for employment and skills across England? Of course, the Learning and Work Institute has a list we’d be happy to talk to them about.

However, not all services and policies can be devolved in all places. This is where the outcome-agreement model helps - setting out a pathway to greater devolution and the support needed to get there.

Some things should still be done at the national level. Apprenticeship standards, for example, and the majority of the tax and benefit system. I remember visiting Canada and being told that one province agreeing to recognise another’s qualifications was a step forward. We clearly don’t want to get to that stage here.

Making the case 

It’s also worth reflecting on how centralised our employment and skills system is compared with many other countries.

When I speak with people in other countries, they are often surprised by how few powers our mayors and local authorities have over employment and skills policy, and how reliant on central government funding they are.

The Netherlands, for example, has much greater local control of employment services - local areas keeping savings in the welfare budget if they do better, but needing to fill the funding gap if they do worse than expected.

Devolution is not an end in itself. The case for it depends on demonstrating how local residents and employers would benefit. After decades of centralisation, local areas now have a chance to prove the difference they can make, and to make the case for further devolution.

Stephen Evans is chief executive of the Learning and Work Institute