The elephant in the room, referring to an issue everybody can see but won't talk about, is one of our most irritating cliches - almost as bad the perfect storm, used to mean trouble on more than one front. So I am employing the elephant's slightly less tired cousin, the gorilla, in a shameless atempt to get a more interesting headline. I want to talk about the politically unmentionable fact of London's local government: its borough system is not fit for purpose.
Local government in London is the responsibility of 32 Boroughs, plus the oddity of the City of London. These boroughs were created in a reorganisation of London's government in 1963. They were amalgamations of a series of much smaller boroughs, and were civil service creations, designed to carefully balance income from rates, the taxes then charged on both business and residential properties - and which funded a large part of each borough's activities. Inner London districts, like Vauxhall, were paired with leafier areas like Streatham. There was little or no attempt made at geographical coherence. They were more or less random collections of London's villages, with boundaries often ignoring even these (Balham is split between Wandsworth and Lambeth, for example; Wimbledon between Wandsworth and Merton, and so on).
Geographically incoherent and designed for a defunct system of local taxes: not a promising start. But that is not the main problem. Fifty years after their creation they are clearly much too small to deliver most of their services efficiently. I have had the same conversation with a number of different people working with local government in a number of different parts of London. They all agree that the boroughs are much to small to be effective, especially in the areas of social services and education - but also in general administration. But politicians won't talk about it.
Two problems are coming together. First is critical mass and economies of scale: there is excessive duplication at senior and administrative levels, and loss of flexibility in the deployment of staff. Then there is geographical incoherence: problems often cross boundaries. I can see this clearly in the area that I have most to do with: education. Even at primary school level: there is a crisis of lack of primary places locally; it is worst at the borough's boundaries - and yet officials are loath to consider new schools where a lot of pupils will come from another borough. The problem gets worse at secondary level - and it is an absolute joke at tertiary - where the colleges boroughs are responsible for service much bigger areas than a single borough.
The efficiency argument is getting a little bit of political attention. The government is encouraging boroughs to merge departments with neighbours. This is being advanced by the tri-borough arrangement of Westminster, Kensington & Chelsea, and Hammersmith & Fulham. But this is ad-hoc and awkward, since the councils themselves are being left intact.
But as well as the boroughs being too small, they are too big. A local sense of place is already quite weak in London. Such as it is works by "village", whose boundaries can be hard to see, but which each have a distinctive centre, often based on an ancient village that was once rural. The borough feels remote and arbitrary by comparison - with residents often feeling that their area is neglected by councillors and officials.
Too big and too small at once. Perhaps there is little surprise that politicians are staying clear of the issue. But it is worth thinking about how it might be organised more sensibly. I want to do this in the opposite way to a classic civil servant: from the bottom up. This starts with the villages. These need to be defined as the basic unit of local government.
My starting point for identifying local villages would be the good old postal districts. These were designed in 1917 by the Post Office, without regard to parish or civil boundaries. But they seem to have known what they were doing, and it is easy to attribute a village name to each district (SW11 = Battersea; N1 = Islington, and so on). They have also become a badge of identity in their own right. This won't always work, and they get a bit dodgy around the edges, but think this gives a broadly viable size. One thing that should be disregarded is current ward boundaries. Wards are designed to be of even population size within a borough - which means they disregard commonly understood village boundaries.
These village should then be grouped into a smaller number of boroughs. Geographical coherence is probably not a realistic objective here - though it makes sense to broadly accept the Thames as a boundary. There should be no fetish about exactly equal size, however. How many? There are two current models. There are six NHS PCT clusters. This would be the radical version, and sort feels like the five boroughs of New York. The less radical alternative would be the 14 constituencies of the Greater London Assembly. This feels like a better balance between Greater London overall, and and the villages.
So how would the two tiers work? The heavy lifting, including all contracts of employment should be at borough level. The main function of the village level would be scrutiny and coordination. But the villages should lead on the critical area of planning. Things like licences for pubs should be decided at this level. Following the example of what goes on outside London, residents could elect councillors at both levels: principal local authorities and local town or parish councils.
But amongst those that have worked with such things there is almost no enthusiasm for separate town council elections. The general level of political apathy may make it difficult to sustain in many areas - and it is a recipe for friction and conflict. A much better idea is for the principal councillors to do both jobs. The borough would form village committees of the councillors elected in each village. This is the neighbourhood committee system which operates successfully in Liberal Democrat boroughs such as Kingston. This places some constraints on size. You would need, say, a minimum of six or seven councillors elected in each village. Let's say there were 120 to 150 villages across London: that would maybe 1,000 councillors. With only six boroughs you would have an average of about 170 councillors in the principal authority. With 14 you would have a more manageable 70 or so. The best way of electing the councils would be one multi-member ward per village, elected by proportional representation (STV or party list). You would simply vary the number of councillors in accordance with the population, without the need to keep reviewing boundaries.
Well that's enough to start with. I haven't even mentioned taxes and debts. But the point now should be to start talking about it. Surely this would be a more efficient and effective way of running the capital? It needs to get on the political agenda. Probably the next step would be for a think tank to take the idea on, and then start drawing in politicians from across the political spectrum. Time to talk to the gorilla.