The latest instalment of our weekly series, in which we use the Centre for Cities’ data tools to crunch some of the numbers on Britain’s cities.

Last week in this space we looked at which British cities had the most start ups. The top three (saved you a click) trend out to be London, Milton Keynes, and – more surprisingly – Northampton, which has pumped a lot of money into subsidising new start ups. This, so it turns out, works.

But it’s one thing to start a business – it’s quite another to make it profitable. So which cities are seeing the most businesses go down the pan?

The map below shows the number of businesses closures per 10,000 population in each city in 2013. (You have to control for population like that, otherwise London would have the most business closures by a very long way, simply by virtue of having the most businesses).

Often these maps of economic metrics show a pretty predictable pattern. Most of the time, southern English cities, York, Edinburgh and Aberdeen do pretty well; pretty much everyone else does less well. Business closures are generally seen as a bad thing, so one might naturally assume there’d be more of them in the latter group.

Not a bit of it though. Some of the cities with the largest number if business closures are in the more affluent bits of the countries.

The top 10…

While those with the smaller number are in the less affluent bit:

…the bottom 10.

The obvious explanation for this is that “business closures per head” is actually a pretty poor proxy for a city’s economic health. That’s partly because a city with more businesses will, all else being equal, have more business closures too. And the raw figures don’t tell us much, even when controlled for population. To get a true sense of what they mean you need to control for the total number of businesses too: after all, the loss of 10 businesses in a city with 1,000 of them is going to hurt less than the loss of one business in a city with three.

So it turns out that – at least in times of economic growth – a high number of business closures is not the disaster one might think. In fact, it correlates really well with the number of business start ups. Really, strikingly well:

The cities at the bottom left of that graph include Hull, Sunderland and Dundee. Those in the top right include Milton Keynes, Brighton and London.

In other words, the key factor lying behind both figures is the total number of businesses in a town. Cities where lots of businesses close can also be boom towns – provided that other businesses are springing up to replace them.

What we’re seeing in action is the painful-to-live-through-but-beloved-of-economists theory of creative destruction. A city where everyone now works for young high tech companies will have seen a lot of businesses close, but it’s almost certainly going to be rich; one where businesses never close, but everyone’s still dependent on ageing heavy manufacturing, is almost certainly going to be poor.

You can like us on on Facebook if you’re into this sort of thing.