As Leeds-based blogger Tom Forth often reminds his readers, Leeds is the largest city in Europe that does not have a tram. That is bad news for Leeds. The Centre for Cities’s report on public transport notes that only 38% of Leed’s population can get to the centre within 30 minutes. If it had a tram, that number would rise to 61% - close to the European average of 67%. 

This is not just, or even primarily, about convenience for people who live in Leeds. It is about economic growth. More people able to get to a city centre means more productivity, as workers find it easier to move jobs, get promoted, get ideas from each other. Competition is stronger, and the best firms win out. That is, literally, one of the most important ways that cities and countries get richer.

So Leeds should have a tram. And yet - even though the outgoing Chancellor, Rachel Reeves, represented Leeds, and was hugely keen on capital investment - nothing ever happens. 

That is why, in a nutshell, we need devolution. We need the mayor of Leeds, or the local council, to have the right to design and build and operate a tram. Or, more likely, to get in partners who can do all of those things. We need them to be able to raise the money to pay for these things, and to take responsibility for their success. And if there is to be a Leeds tram, Leeds should choose the routes, not London. 

For sure, national government will always have to provide some sort of backstop level of support. When something like Covid-19 hits, only national government is big enough to step in - as it did for Crossrail, when costs rose because of covid-induced delays. But, by and large, places like Leeds should decide whether they get a tram, how extensive it should be, what fares it should charge, and so on. 

If local people don’t like the proposal, they can vote for a different mayor or for another party. That is how local democracy works.

We can, therefore, judge the success of devolution by whether ‘things’ happen. There are any number of ‘things’ that could be devolved to local areas. Sometimes areas will choose to use their freedoms to make changes, sometimes they will not. That will be up to local people, and the people that they elect. 

If you travelled to Hull before mobile phones were common you would notice the absence of red telephone boxes, because in Hull the phone boxes were white. For reasons lost in the mists of time, telephone lines in Hull were owned by the council, not by British Telecom, which, back in the day, was a national monopoly and part of the Post Office. 

While no mayor or local council would set out today to create a local fixed line telephone service, they should be allowed to if they wish to. I can imagine a local mayor sorting out 5G signals in their area by building a lot of phone masts. They could recoup the money from service changes - literally, charging Vodafone, EE and so on when a call is routed through these masts. Or they could make a loss on the grounds that a great 5G service is what local people want, makes the area attractive to business and is a route to prosperity. 

A really radical government could devolve a lot. Rather than force all areas of the UK to have a maximum £2 bus fare, the UK government could have given cash grants to local areas. Some might choose to have a £2 bus fare. Others would have used the money to run buses earlier in the day, or later in the evenings. Others might choose a more frequent service, or newer buses. Other areas might decide to spend the money creating new bus lanes, so that the buses are faster rather than cheaper. Perhaps in some areas that is a better way to get more people to use buses. Devolution is about letting local places decide what is best for them. Other areas might use the money to build more cycle lanes, if they think that is a better way to encourage more people to visit the centre of towns, or to leave their cars at home. Others might want to offer a subsidised e-bike scheme. Still others might decide either that the buses are fine, or that buses are fundamentally ill-suited to their area. In such cases the council might want to spend the money improving the town centre, improving adult skills, or simply reducing council tax so that people are not as hard pressed. The point of devolution is that local people get to decide. 

Let me say it again: a really radical government could devolve a lot. Take education, for example. A really radical government could allow a mayor or local council to raise (or lower) the age at which children start school. Many countries have their children starting later than us. An authority that did this should be able to bank the savings, or at very least to use them to better fund early years education. That radical government could also allow mayors and local authorities to require their schools to offer different qualifications altogether. Schools in Leicester could offer the International Baccalaureate, while schools in Newcastle could offer Scottish Highers. A local authority could decide that Ofsted does more harm than good, and keep them out of schools in their area. Alternatively it could opt out of the phonics check, or Key Stage Two testing. It could require all children to take two vocational courses to GCSE level, or require a modern foreign language, or…well, the list is endless.

Now I happen to think that most of these ideas are not sensible, and will harm children’s life chances. That is my reading of the evidence - and I would point to PISA evidence for England relative to Scotland and Wales to defend my position. But, bluntly, why should the people of Huddersfield take any notice of what I think? If they think that the current system of education is failing their children, a government committed to devolution should allow them to change that system of education.

This then, is the crux. Devolution cannot guarantee better outcomes, and certainly not better outcomes everywhere. Education has been devolved to Wales, Scotland and Northern Ireland, and as the PISA results show, this has not led to better results in these areas. 

But even when it does not, devolution tells us something useful, because it gives us good evidence of what works and what does not. I gain if your area chooses to do something differently. If it works, my area can copy yours. If it does not, my area knows not to copy you. 

That is why the one thing I would insist on is that evidence is collected as to what works. It is a travesty that we don’t have comparable data for the number of young people who are not in employment, education or training in Scotland. We have it for all the regions of England, and for Wales, but the Scottish government refuses to publish comparable data. That means that none of the other countries of the United Kingdom can learn from Scotland, and Scotland cannot benchmark itself against the rest of the UK. Of course it is politically embarrassing when one country does worse, but that embarrassment is a price politicians should be forced to pay when their ideas do not work out.

Moving power to local levels means moving taxes to local levels. Otherwise the mayor can blame everything on lack of money from national government. Mayors should be people who sort out local problems, not the person whose job it is to complain most loudly, and get money out of Westminster most effectively. 

There are often calls for local income taxes. These can work - the US has them, after all - but they are not the obvious taxes to devolve. They work very well in rich places, where a very small tax will raise a lot of money. They work much less well in poorer areas, where a massive tax hike would be needed to raise a decent sum of money. Indeed, there is a serious risk that a local income tax would lead to the sort of unhelpful outcomes we have seen in the US, whereby a high tax rate in one area leads people to move out, which destroys the local tax base, and local services, leading more people to move out, setting off a genuine spiral of decline. 

Broadly speaking, local taxes should be regressive and place-based, and national taxes progressive and need not be place-based. Only by making local taxes regressive and place based can we ensure that all local areas have similar tax bases, and face similar choice sets, and have stable revenue streams. As an example, devolving the revenue from VAT on domestic energy would give a relatively similar amount of revenue per head to all areas. Although motorists can choose where to fill up, a local petrol tax would probably work reasonably well, at least for larger areas, and at least outside London. But a local tax on buying a car would lead car showrooms to move out of some authorities, if the tax was too high. Equally, allowing a council to levy air passenger duty would be wonderful for Hillingdon (Heathrow), Crawley (Gatwick) and Uttlesford (Stansted), but would hardly be a sustainable basis for local council finance more widely. Other taxes that would work well would include licences - for example, allowing a local authority to charge more than costs for licences for off-licences, fried chicken shops and so on. 

As ever, Britain would do better not to try to invent all this, but to simply learn from other countries. The OECD has collated data on subnational government revenue, which would identify the places that a radical government committed to extensive devolution should look to emulate. It has also written about how to combat unhelpful tax competition. As a general rule of thumb, if it works in other countries that are broadly similar to ours, it will probably work here.

Democracy works best when people have a genuine choice as to who governs them. That is definitely the case at national level, where the governing party changes often enough that all politicians are aware that there are consequences of making bad decisions. In contrast this is not always the case at local level, where in some cases the same party has been in power in an area for many, many years without any obvious connection to their effectiveness or success locally. The method we use to elect local councillors - typically ‘multi-member first past the post’ - is the system most likely to give an overwhelming majority - and sometimes all of the seats - to a party that wins a relatively small proportion of the votes. Any serious move towards devolving significantly greater levels of control to local areas should be accompanied by moves to revitalise local democracy, including reforms to the voting system that make ‘one-party states’ much less common. 

Devolution can work. It works in English speaking countries such as the US, Canada and Australia. It works in nearby countries that are similar to us, including most obviously France and Germany. There is absolutely no reason why devolution cannot work here. There can be no guarantee, but Britain probably will be better governed and richer if this happens. Reversing decades of centralisation would certainly be a legacy for any prime minister seeking a place in the history books.