Premier league poverty, 2015

Over the past decade I've spent a lot of time looking at patterns of deprivation across the UK. One thing I've often noticed is the way football grounds regularly appear in the very poorest neighbourhoods. I've blogged on this topic a few times in the past, most notably in 2012 when I looked at the location of English Premier League grounds in relation to the deprivation level of their areas. I also noticed this in relation to the Scottish Index of Multiple Deprivation when looking at the East End of Glasgow in 2009. Given the history of football, its industrial working class origins, the development of British cities, and land values (to name just a few factors), none of this should be a surprise. But since a new English deprivation dataset was released in September, I wanted to revisit the topic and make a few maps, just to see if anything has changed. That's what I've done here - one map showing the location of each Premier League ground and the deprivation level of the area it sits in - and its wider neighbourhood. Further explanation follows below.

Note the blue area to the north east - now Highbury Square

Most areas in the neighbourhood are in the 20% most deprived

Bournemouth is the one big exception - it's at the opposite end of the scale

Stamford Bridge is in a much more mixed area than most

Selhurst Park sits right beside some more deprived areas

Goodison and Anfield look very similar  - only about half a mile apart

The wider area of Leicester's ground is more mixed

This is quite typical of much of north Liverpool

Manchester City play in the most deprived area of any top flight team

Manchester United are situated in a more mixed land use area

Newcastle United's pitch is split between areas - I've based this on majority area

Norwich City's ground is also in a more mixed neighbourhood

St Mary's is situated in one of the city's most deprived zones

Stoke also play in quite a deprived area - though there's more variation nearby

Like many newer stadiums, this ground is in a slightly more mixed area

The Welsh deprivation dataset is used here - but similar story to be told

Post-riots, much has been said of regeneration in this area of London

Watford play in a much less deprived locality

West Brom's ground is in the most deprived decile of England

Another ground split between areas - but still more deprived than not

What does all this tell us?

The most obvious thing to emerge from this simple mapping exercise is that more than half of all Premier League grounds are located in areas among the 20% most deprived in the country, but a good few are not. Two in particular - Bournemouth and Watford - are in much less deprived areas. Nonetheless, if you scroll through the maps quickly, the main colour you'll see is red (for the 20% most deprived). When I see stories in the news about the ability of sport to tackle deprivation, I'm generally all for it, but then sometimes I make a mental comparison between the wage bills of some teams and the neighbourhoods they're located in and I think we've barely scratched the surface of what's possible when we talk of the potential for elite sport to help transform poorer areas. Post-Olympics, this has kind of been forgotten. Having said this, it is good to see that the Premier League and FA's Football Foundation provides money for grass roots development in the most deprived areas as defined by this very same dataset. 

What does it not tell us?

Quite a lot, and I wouldn't want anyone to think that I've done this to pick on any one team. I'm just curious about the relationship between these football grounds and underlying patterns of deprivation because when I look at the data as I map it, I often notice the stadia. It doesn't tell us anything about cause and effect, whether teams are trying to do anything to boost the fortunes of their local areas or what the areas themselves are like to live in. If you want to know more about the underlying data, read this briefing from the Government. Does having a Premier League football team in your area make you poor? Of course not. 

Some of the grounds look the wrong shape - why?

I used building footprint data from the Ordnance Survey in the maps above and the shapes of the grounds are as they were in the original dataset - with the exception of Vicarage Road, which for some reason wasn't enclosed on one side so I made my own version. I've just added a little glow around each ground to make it stand out and then added in the footprints of all other buildings in the wider neighbourhood to help people identify nearby features and roads.

What about when a ground is split between areas?

I could have taken the average deprivation rank here and used that figure but instead I chose to use the deprivation rank of the area that the majority of the playing surface was located in. This was only really an issue for Arsenal, Newcastle and West Ham - and only really notable in Newcastle. 

Explain that 'deprivation percentile' thing again please

In England, there are 32,844 areas known as Lower Super Output Areas. These LSOAs are small areas which the government use to report all kinds of statistics, including Census data. When they publish their Indices of Deprivation, they give each one of the 32,844 areas a rank, from 1 (most deprived in England) to 32,844 (least deprived in England). Therefore, it's a relative measure that allows us to compare one area with another, all across the country. The data are often split into five or ten chunks (quintiles or deciles) for reporting purposes but here I've decided to use 'percentiles' as it's more precise. If an area is in percentile 5, it's among the 5% most deprived in England, and so on. If it's in percentile 95 (like Bournemouth's ground) then we can say it's not very deprived at all and actually highly likely to be a very affluent area. In the case of Swansea City, I've used Welsh deprivation data from 2014. This classifies places in almost exactly the same way, although there are 1,909 areas in Wales rather than 32,844. These areas have an average population of around 1,600.

Isn't this all just pointless area classification?

You might think so, but the Government use these Indices to make all sorts of important decisions, in healthcare and education for example. If you're in an area classified as being among the 20% most deprived, for example, you might find that you're eligible for some kind of funding - there are loads of examples of uses, with sport being one of many. You can find quite a few other examples in section 1.4 (p. 8) of this report. We must also remember that not all people living in areas classified as 'deprived' or 'not deprived' match that description - this dataset classifies areas not people.

When are you going to expand this to include my team?

I'm not planning to, but I'm sure it would be even more interesting than the Premier League.


Curiosities
On all the maps, north is up so I couldn't help notice that Manchester United seem to be the only team playing on an east to west pitch. I'm guessing most grounds don't do this so that they can avoid the setting sun problem - and in fact Old Trafford cricket ground rotated their pitch 90 degrees to avoid this problem in 2010. Shades of blue - representing the 40% least deprived areas - appear on only 7 of the maps, and only two grounds are in such areas. Red (20% most deprived areas) appear on 19 maps - only Watford is the exception. The maps for Everton, Man City, Tottenham and West Brom are entirely red - which indicates that these grounds and surrounding areas (a few hundred metres in each direction) are within wider areas classified as the most deprived in England. The very most deprived areas to appear on any of the maps are ranked 24 (beside Goodison) and 29 (beside Anfield). 

Which team do you support?

ICTFC, of course. But not very enthusiastically. 

London's 100 Poorest Areas

The theme for this post follows on from some work I did recently which looked at the increasing level of deprivation in Outer London, as reported in the Guardian a couple of weeks ago - see also the maps on the Guardian datablog. It's true that housing market pressures (among other things) are helping to push poverty away from Inner London but the majority of London's poorest are still within the inner city, and Tower Hamlets, Hackney and Newham in particular. In order to provide a clearer picture of where exactly the very poorest (most deprived) areas are, I produced a map and animation of London's 100 most deprived areas. 

Animated version

The results are not at all surprising. Of the 100 most deprived LSOAs in London, according to the 2010 Indices of Deprivation, the Borough with the most areas is Tower Hamlets (18), followed by Hackney (17), Newham (13), Haringey (12) and Brent (10). Deprivation is increasing in Outer London relative to the past and I was particularly taken today by the news that Newham Council were looking to re-house poorer residents in Stoke (more than 150 miles away). That's taking the suburbanisation of poverty a bit far! 

Some work I did with Ed Ferrari last year looked at residential mobility patterns amongst the richest and poorest sectors of the population in England. What we found surprised us at the time. Our analysis showed that it was often the poorest areas which were associated with the longest residential moves. We didn't have enough fine-grained data to make more of this but it was an interesting insight into what might be happening more broadly.

Taking the 100 most deprived locations is of course arbitrary but the point here is that despite debates about whether 'the poor' should live in 'rich areas' the fact is that many of the poorest people living in London are in areas which in recent years have changed considerably so that they are now experiencing very high demand, inflated rents and severe socio-economic inequalities. This represents an intensification of existing processes rather than something entirely new but it does mean that policy makers need to think carefully about what to do about it...

Divided Cities?

There has been a lot of talk recently about the links between deprivation and the recent riots in England (e.g. this from the Guardian), and since I'm interested in the issue I thought I would do some very basic analysis to look at the spatial divisions of deprivation in English cities. To do this I took London, the eight core cities plus Bradford, Coventry, Hull and Leicester and mapped the 10% most (in red) and 10% least (in blue) deprived (IMD 2010) areas on one large map graphic. The results are shown below - click the image to enlarge.

Some interesting comparisons can be made from this image. In Hull, Leicester, Manchester and Nottingham there are no areas amongst England's 10% most deprived and in Liverpool there is only one. In Sheffield there is a clear NE/SW split in terms of the location of the most and least deprived and in Manchester, Birmingham and Liverpool the number of areas in the most deprived 10% is quite high.

Not much else to say now except that I think in studies of deprivation we should perhaps be more concerned with inequalities and how they manifest themselves spatially.

Indices of Deprivation 2010

The English Indices of Deprivation 2010 were published this morning. Previous Indices were published in 2000, 2004 and 2007. I've decided to do a few posts on this because I find it very interesting and it does have real implications in relation to policy. In this post, I've done some basic mapping and analysis (click on any map to view it full size).

A quick snapshot first: the most deprived LSOA is in Tendring, Essex. It used to be in Liverpool. Overall, Liverpool, Middlesbrough, Manchester, Knowsley, Hull, Hackney and Tower Hamlets remain the most deprived local authorities. So, no major surprises.

Overview for the whole of England, with a list of top ten most deprived LSOAs.


Overview cartogram for England, with a list of top ten most deprived LSOAs.


Detailed map showing location of the most deprived LSOA in England.


A zoomable google map version of the above, for good measure.

View Tendring 018A in a larger map

A more focused map of Merseyside, showing Liverpool, Sefton and Knowsley.

Note: The Indices of Deprivation 2010 are produced for small areas called LSOAs. These lower layer super output areas are not 'neighbourhoods' in a formal sense but they often have a high degree of internal similarity. The average population of an LSOA is about 1500 and in total there are 32,482 of them in England. The data used in the Indices are mostly from 2008 but you can read more about this in the technical report.