Alasdair Rae
Aug 9 · 8 min read

In this research note Alasdair Rae and Elvis Nyanzu from the University of Sheffield discuss inequality and present some examples from their ongoing Atlas of Inequality project, funded by the Nuffield Foundation. They’ll publish their findings in November 2019 alongside a set of inequality data and maps for all local areas across England. Their hope is that it will lead to a more nuanced, geographical understanding of inequality and that it will make a contribution to the debate about how we deal with inequalities on the ground.

Inequality has many guises

Did you know that the most expensive property in England in 2018 sold for £160 million? You did? Okay then, did you know that with this money you could have bought all 800 houses sold in Merthyr Tydfil and still have £65 million to spare? This kind of statistic is merely an anecdote, but it’s a powerful reminder of the extremes of wealth that exist in modern Britain. Indeed, one could argue such examples are partly to blame for the perception of inequality as a serious social ill.

Source: Pixabay

Yet in our study we don’t approach inequalities through the house price looking glass, however tempting it might be. Instead we focus on overall income distributions, deprivation ratios and spatial patterns of inequality. By taking a multi-faceted geographical approach, rather than relying solely on a single indicator (such as the Gini coefficient), we hope it will be possible to arrive at a more rounded understanding of inequality on the ground. After all, inequality has many facets and not all of these can be captured with a single figure.

This isn’t new: so what would Plato say?

Owing to a steady stream of books and newspaper articles, inequality can seem like a particularly contemporary phenomenon, but this is far from the truth. Around 2,400 years ago, Plato wrote The Republic and in this Socratic dialogue he made an observation that seems particularly pertinent in discussions about inequality today:

“Any city, however small, is in fact divided into two, one the city of the poor, the other of the rich.”

Plato, The Republic (Book IV), 380 BC

If he were alive now, Plato probably wouldn’t have much trouble identifying ‘rich’ or ‘poor’ areas in English cities, particularly in places like Sheffield. But we also suspect he wouldn’t find himself embroiled in technical arguments about Gini coefficients or Theil indices. However, since we’re not actually able to speak for one of the pivotal figures in the development of the Western intellectual tradition, this is just idle speculation. The point here is that inequality has been recognised from antiquity as a social problem worth talking about.

More recently, inequality has become a prominent feature of public discourse, but one that is often sidetracked by arguments about measurement and context. By producing an English Atlas of Inequality, we hope to bring a slightly different perspective to the table, and one that is inherently geographical and not wedded to a single metric. We see the geographical element as being particularly important since, as in Plato’s day, the income divide is also often a spatial divide (see map below).

Sheffield: blue areas are among the 10% least deprived in England, red areas among the 10% most deprived

Disraeli’s two nations: Hart and Hartlepool?

When they’re not citing Plato, inequality scholars refer instead to Disraeli and his famous quote from Sybil, or The Two Nations, where the young aristocrat Charles Egremont talks of ‘two nations between whom there is no intercourse and no sympathy … the rich and the poor’. This, as you may know, is where you get the notion of ‘One-nation Conservatism’ from and in the context of the divisions currently rife within British society it’s not too much of a leap to think that there are two nations.

We wouldn’t go this far, and we think if reality met hyperbole they probably wouldn’t recognise each other. That is, new extremes of public discourse have led to a situation where society is increasingly framed in oppositional terms, with little room for the nuance and subtlety that undoubtedly exists in between.

Perhaps we can blame social media for that, but perhaps geographical separation also plays a part when the lives of people at opposite ends of the income spectrum rarely cross paths.

Two English local authorities serve as a useful example of this. On an alphabetical list of English local authorities, Hart and Hartlepool come one after the other so when analysing our data we often notice the contrast between these two areas. Hart is almost equidistant between central London and Southampton and is named after the River Hart in Hampshire in the south of England, whereas Hartlepool is a former steel-making powerhouse 300 miles to the north.

Almost three quarters of Hart falls within the least deprived 10% of England as a whole, as measured by the 2015 Indices of Deprivation. A third of Hartlepool’s neighbourhoods, by contrast, sit within the most deprived 10% of England on the same measure. This kind of contrast is neither uncommon nor surprising but it does help explain the kinds of narrative around ‘two nations’ and ‘the left behind’ that have recently gained prominence in national discourse (terms we find somewhat unhelpful in moving the debate forward).

Such themes, and many more, were deftly captured by Andrés Rodríguez-Pose in his recent essay on ‘The Revenge of Places that Don’t Matter’. Unlike much academic commentary, Rodríguez-Pose also explains what he thinks we can do about it and is clear that “doing nothing is not an option, as the territorial inequalities at the root of the problem are likely to continue increasing, further stirring social, political, and economic tensions”.

It is tempting to think that there are two Englands and indeed by looking at Hart and Hartlepool it is easy to see things in this way, but these somewhat stereotypical north-south examples in part overlook the arbitrary nature of the boundaries we use to measure such phenomena and the fact that ‘deprivation’ is only one way to understand inequality. That’s why it is only one element of our study, albeit an important one.

Which places are most unequal?

Answering the question of which parts of England are ‘most deprived’ is straightforward, since we can refer to official small area statistics and compare one area with another. Yet this is just one element of the inequality equation. Answering the question of which places are most unequal is more difficult but in this project we’ve attempted to do so using different boundaries and different measures. We’ll discuss one such approach now, and publish more in our Atlas later in 2019.

When looking at inequalities that are mainly about income, we think it makes sense to use an income-related geography. For that reason, we use travel to work area (TTWA) boundaries in our research as they’re a reasonable approximation for labour market areas.

Alongside this, we’ve used three measures to understand inequality. The first is a simple ‘20:20’ ratio looking at the proportion of each area among the most and least deprived 20% in England. The second is a local gini coefficient (derived using equivalised ONS household-level income data). The third involves some spatial statistical analysis to derive what might be thought of as a ‘cheek-by-jowl’ index, to understand the extent to which rich and poor areas are dispersed, or cluster together.

We’ll look at the 20:20 ratio now, as a measure of inequality that highlights the proportion of places classified among the most or least deprived 20% in England as a whole. Some of the areas on the list below have far more deprived areas than less deprived ones (e.g. Hull, Luton), whereas others are the opposite (e.g. Reading, High Wycombe and Aylesbury), yet in all cases there is significant inequality between what Plato might have referred to as ‘the rich’ and ‘the poor’.

Source: Indices of Deprivation 2015, MHCLG; calculations by the authors

On the map of High Wycombe and Aylesbury below you can see one particular type of inequality, where those living in less deprived areas within a wider region significantly outnumber those in the poorest (although many of the blue areas are sparsely populated). This measure tells us something about local economic imbalances, and we think it serves as a useful contextual indicator of localised inequalities, particularly when used alongside our other measures. This gives us a more rounded view of localised inequalities across England and, we hope, sheds new light on the topic.

Map showing areas among England’s most and least deprived 20%, within High Wycombe and Aylesbury TTWA

Complicating factors: “not all poor people live in poor areas”

Before we conclude, it is important to mention one potential complicating factor; namely, the idea that ‘not all poor people live in poor areas’ Indeed, it is sometimes said that most poor people do not live in ‘poor’ areas (e.g. see p.1341 of this paper).

It’s important for us to consider this in our work because most of the data we use relate to areas rather than individuals. The only exception is the individual income data we use to generate our local gini coefficients, where we’ve followed the same approach as Paul Swinney and his team at the Centre for Cities.

The “not all poor people live in poor areas” claim is often repeated by critics of socio-spatial analyses which take areas as their base unit, but it is hard to find evidence to either support or reject it, so we are investigating it further. What we have come up with as an initial check is to look at individual areas by deprivation decile and then, within these categories, look at the proportion of households in different income bands, published as experimental research outputs by the Office for National Statistics. They are not an indicator of poverty or living standards but they do help shed further light on the variation in incomes evident within different types of areas.

You can see the results in the chart below, which relates to income from Pay As You Earn (PAYE) and benefits in the tax year ending in 2016. The precise breakdown in each stacked bar is of less relevance than the fact that each one contains many different colours, indicating a level of income diversity that is usually hidden from view by aggregate area measures.

This provides some new insight into the diversity of lived experiences that are likely to exist within what are often treated as canonical data categories. We shall be investigating this issue further as we continue our research.

Household income by individual deprivation decile for England, 2016 ( on the x-axis, 1 = most deprived decile, 10 = least deprived)

We’re working on a new Atlas of Inequality for England

Our work to date has helped us to move on both conceptually and empirically from simply identifying the ‘poorest’, ‘richest’ or ‘most deprived’ areas of England. It has also helped us understand more fully the implications of using different boundaries to answer the same question.

We use the example of travel to work areas above, but we will also report our results for local authorities and parliamentary constituencies when we publish our English Atlas of Inequality later in 2019. When we do, we’ll focus on the following three questions.

  • How is inequality best measured at the local level?
  • Do more equal areas have better overall outcomes?
  • Which areas of England are most unequal, and how does this affect the life chances of poorer individuals?

Alasdair Rae

Written by

Stats, maps, cities, regions, housing, transport, gifs & some QGIS stuff.

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade