Does this map actually show every European town with over 1,000 residents?

Mmmm, misleading. Image: Sp33d3h/Reddit.

Sometimes in my travels around the backwoods of the internet I spot a map so pleasing, one that does such a perfect job of telling a story, that it goes straight to the top of the list of things I plan to pontificate about as soon as I find a spare moment.

And then I start to wonder whether perhaps it might just be a bit too good to be true.

So it was with this beauty, posted to Reddit by user Sp33d3h. It uses a single red dot to show every town or city in Europe with a population of over 1,000. You can instantly see which bits of the UK are heavily populated, and which are relatively rural. You can see the way Europe's population thins out the further you head north or east. You can see the Alps.

It's lovely.

Click to expand.

It's probably not right, though.

For one thing, the boundaries between countries are often just a little too well demarcated. You can see at a glance where Denmark ends and Germany begins; you can also spot the boundaries between Bulgaria and Romania, or Poland and Slovakia.

Rough national boundaries in blue.

Now, it's possible that there's some geological feature demarcating the boundary which means that urbanisation becomes much heavier on one side than the other – but the fact those boundaries (most of which are relatively recent inventions) are visible over such long distances suggests something else is going on. It looks like me like there’s a disconnect in the data – like it’s being collected in different ways in different countries.

Which, if you think about it, it obviously would be: literally nobody has the resources to go round counting all the settlements of 1,000 people or more across an entire continent. It's all but certain that this map is collating datasets collected by other people, and national boundaries are the most likely place for one dataset to stop and the next begin.

One of the commenters on Reddit who's examined the map on the Harvard website that provided the data for this one speculates that what we're actually looking at is municipal or administrative boundaries. Except if we zoom in it's not even clear that's it, because here's what you get in London:

Nope, no idea. Image: Harvard WorldMap.

I am reasonably familiar with all sorts of ways of chopping my city up into bits, and that doesn't look like any of them. I have literally no idea what these units are. (Also, it's not immediately obvious why the Greater London conurbation should really be a few dozen settlements rather than one big one.)

So – as pleasing as this map is it probably isn't anything as useful as every European settlement of over 1,000 people.

That doesn't mean we can't tell anything useful from the map. The way the dots are distributed within countries is probably quite telling. So we can see, for example, that the most rural parts of Great Britain are in mid Wales, Scotland outside its central belt, and the far north west of England:

We can see that the fringes of France and Spain are generally more populated than the middle:

And that Scandinavia empties out, the further north you get:

But internationally, I fear, this map tells us little – or rather, we can't tell when it is telling us something, and when it's just a quirk of the data.

As much as I'd love to see a map which actually did what the one at the top of this post claims to, I fear that no such data set exists. I mean, how would you even begin to count every settlement that small?

For what it’s worth, here's a map showing population density across Europe, using roughly county-sized lumps. Enjoy.

Image: DBachmann/Wikimedia Commons.

Jonn Elledge is the editor of CityMetric. He is on Twitter as @jonnelledge and also has a Facebook page now for some reason.

Want more of this stuff? Follow CityMetric on Twitter or Facebook.



Is Britain’s housing crisis a myth?

Council housing in Lambeth, south London. Image: Getty.

I’ve been banging on about the need for Britain to build more houses for so long that I can no longer remember how or when it started. But at some point over the last few years, the need to build more homes has become My Thing. People ask me to speak at housing events, or @ me into arguments they’re having on Twitter on a Sunday morning in the hope I’ll help them out. You can even buy a me-inspired “Build More Bloody Houses” t-shirt.

It’s thus with trepidation about the damage I’m about to do to my #personal #brand that I ask:

Does Britain actually have enough houses? Is it possible I’ve been wrong all this time?

This question has been niggling away at me for some time. As far back as 2015, certain right-wing economists were publishing blogs claiming that the housing crisis was actually a myth. Generally the people who wrote those have taken similarly reality-resistant positions on all sorts of other things, so I wasn’t too worried.

But then, similar arguments started to appear from more credible sources. And today, the Financial Times published an excellent essay on the subject under the headline: “Hammond’s housebuilding budget fix will not repair market”.

All these articles draw on the data to make similar arguments: that the number of new homes built has consistently been larger than the number of new households; that focusing on new home numbers alone is misleading, and we should look at net supply; and that the real villain of the piece is the financialisation of housing, in which the old and rich have poured capital into housing for investment reasons, thus bidding up prices.

In other words, the data seems to suggest we don’t need to build vast numbers of houses at all. Have I been living a lie?

Well, the people who’ve been making this argument are by and large very clever economists trawling through the data, whereas I, by contrast, am a jumped-up internet troll with a blog. And I’m not dismissing the argument that the housing crisis is not entirely about supply of homes, but also about supply of money: it feels pretty clear to me that financialisation is a big factor in getting us into this mess.

Nonetheless, for three reasons, I stand by my belief that there is housing crisis, that it is in large part one of supply, and consequently that building more houses is still a big part of the solution.

Firstly I’m not sold on some of the data – or rather, on the interpretation of it. “There is no housing crisis!” takes tend to go big on household formation figures, and the fact they’ve consistently run behind dwelling numbers. Well, they would, wouldn’t they? By definition you can’t form a household if you don’t have a house.

So “a household” is not a useful measure. It doesn’t tell you if everyone can afford their own space, or whether they are being forced to bunk up with friends or family. In the latter situation, there is still a housing crisis, whatever the household formation figures say. And there is plenty of anecdotal evidence to suggest that’s the one we’re living in.

In the same way I’m not quite convinced that average rents is a useful number. Sure, it’s reassuring – and surprising – to know they have grown slower than general prices (although not in London). But all that figure tells you is the price being paid: it doesn’t tell you what is being purchased for that payment. A world in which renters each have their own property may have higher rents than one in which everyone gets one room in an over-crowded shared flat. It’s still the latter which better fits the label “housing crisis”.

Secondly, I’m entirely prepared to believe we’ve been building enough homes in this country to meet housing demand in the aggregate: there are parts of the country where housing is still strikingly affordable.

But that’s no use, because we don’t live in an aggregate UK: we live and work in specific places. Housing demand from one city can be met by building in another, because commuting is a thing – but that’s not always great for quality of life, and more to the point there are limits on how far we can realistically take it. It’s little comfort that Barnsley is building more than enough homes, when the shortage is most acute in Oxford.

So: perhaps there is no national housing crisis. That doesn’t mean there is not a housing crisis, in the sense that large numbers of people cannot access affordable housing in a place convenient for their place of work. National targets are not always helpful.

Thirdly, at risk of going all “anecdote trumps data”, the argument that there is no housing crisis – that, even if young people are priced out of buying by low interest rates, we have enough homes, and rents are reasonable – just doesn’t seem to fit with the lived experience reported by basically every millennial I’ve ever met. Witness the gentrification of previously unfashionable areas, or the gradual takeover of council estates by private renters in their 20s. 

A growing share of the population aren’t just whining about being priced out of ownership: they actively feel that housing costs are crushing them. Perhaps that’s because rents have risen relative to wages; perhaps it’s because there’s something that the data isn’t capturing. But either way, that, to me, sounds like a housing crisis.

To come back to our original question – will building more houses make this better?

Well, it depends where. National targets met by building vast numbers of homes in cities that don’t need them probably won’t make a dent in the places where the crisis is felt. But I still struggle to see how building more homes in, say, Oxford wouldn’t improve the lot of those at the sharp end there: either bringing rents down, or meaning you get more for your money.

There is a housing crisis. It is not a myth. Building more houses may not be sufficient to solve it – but that doesn’t meant it isn’t necessary.

Jonn Elledge is the editor of CityMetric. He is on Twitter as @jonnelledge and also has a Facebook page now for some reason. 

Want more of this stuff? Follow CityMetric on Twitter or Facebook